Elbow Dislocations

                              Elbow Dislocations  

                                             Dr. KS Dhillon

     

Anatomy and biomechanics of the elbow

There are three joints at the elbow:

• Ulnohumeral joint
• Radiocapitellar joint
• Proximal radioulnar joint

Ulnohumeral joint

The ulnohumeral joint is formed by the articulation between the spool shaped distal medial flare of the humerus called the trochlear and the trochlear notch of the proximal ulna formed by the olecranon and coronoid parts of the ulna. It is a hinge type of joint. Forty percent of the axial load in an extended elbow goes through this joint.

Radiocapitellar joint

The radiocapitellar joint is formed by the the capitellum which is situated on the distal lateral flare of the humerus and the head of the radius. It is a pivot type of joint. Sixty percent of the axial load, with elbow in extension, passes through this joint. The radial head is covered by cartilage for about 240 degrees. The lateral 120 degrees contains no cartilage and this information  useful when internal fixation needs to be carried out.

Proximal radioulnar joint

The proximal/superior radioulnar joint is formed by the head of the radius which articulates with the radial notch of the proximal ulna and the joint shares the capsule of the elbow joint. The two bones are held together by the annular ligament which is attached to the anterior and posterior margins of the radial notch of the ulna. The annular ligament circles round the head and neck of the radius and is devoid of ligamentous attachments which enables the radius to rotate freely inside the annular ligament. It is a pivot type of joint and it allows supination and pronation movements of the forearm.

The elbow joint consists of two types of articulation, and it allows two types of motion. The  ulnohumeral articulation is a hinge joint and it allows flexion and extension, whereas the radiohumeral and proximal radioulnar joint are pivot joints which allows axial rotation. Stability of the elbow joint is provided by the bony articulations and the medial and lateral collateral ligaments.

The primary static stabilizer of the elbow is the ulnohumeral joint (coronoid), medial (ulnar) collateral ligament (MCL) and the lateral collateral ligament complex (LCL).
The ulnohumeral stability is provided its bony counters and the coronoid process. A 50% or more loss of coronoid height results in elbow instability. The MCL is composed of the anterior, posterior and transverse bundles. It arises from the posterior medial epicondyle and inserts on the sublime tubercle of medial coronoid process.

The LCL consists of the radial collateral ligament (RCL), lateral ulnar collateral ligament (LUCL), accessory collateral ligament and the annular ligament. The radial collateral ligament (RCL) extends from the lateral epicondyle to the annular ligament deep to the common extensor tendon. The lateral ulnar collateral ligament (LUCL) extends from the lateral epicondyle to the supinator crest on the ulna. The annular ligament (AL), extends from the posterior to the anterior margins of radial notch on the ulna. It encircles the head of radius and holds it against the radial notch of ulna. The accessory lateral collateral ligament (ALCL) extends from the inferior margin of the annular ligament to the supinator crest.

The secondary static stabilizers are the radiocapitellar joint, the capsule and the attachment of the flexor and extensor tendons (biceps, brachialis, brachioradialis and triceps).
The dynamic stabilizers (muscles crossing the joint) of the elbow includes the anconeus, brachialis, triceps and the biceps.

Classification of elbow dislocation

Dislocations of the elbow can be anatomically classified according to the position of the radius and ulna in relation to the humerus after injury. Based on this there are five types of elbow dislocations :

• Posterior (most common)
• Anterior
• Medial
• Lateral
• Divergent (radius and ulna are dislocated in different directions in relation to humerus)

Complex or simple

Depending on the complexity of the dislocation, the dislocations can be class as simple or complex. Simple dislocations are those where a dislocation occurs without associated fractures and complex dislocations are those where an associated fracture or fractures occur. About 50-60% of  dislocations are simple without associated fractures.

There are three common patterns of complex elbow fracture-dislocations, the trans-olecranon fracture-dislocation, the terrible triad injury, and anteromedial coronoid fractures associated with varus posteromedial instability.

Complex elbow dislocation

Initially evaluation

Clinical examination is carried out to rule out open fractures, neurovascular compromise, and associated injuries. Plain X rays are usually sufficient to diagnose complex elbow fracture-dislocations. A closed reduction is carried out if the elbow is dislocated or the limb is grossly deformed. A CT scan of the elbow is than done to assess the fractures and to guide preoperative planning and treatment.

1.Trans-olecranon fracture-dislocations

Trans-olecranon fracture-dislocations, which result from an axial loading injury, are characterized by the disruption of the ulnohumeral joint with  anterior displacement of the radial head relative to the capitellum. Usually there is a complex, comminuted fracture of the proximal ulna, though a simple or an oblique fracture can also occur [1]. Fractures of the coronoid are also commonly associated with these type of injuries and often the fracture  involves more than 50% of the coronoid height [2]. The radial head can also be fractured [1,3,4]. Coronoid fractures can occur concomitantly as well [5, 7, 8]. The collateral ligaments usually remain intact [1,5].

The fractures are treated by internal fixation via the posterior approach. The coronoid or radial head fracture can approached through the exposure afforded by the olecranon fracture. The olecranon fracture can be stabilized with a plate or a tension band wire. The tension band wire fixation has a higher failure rates [3,4]. Occasional additional medial or lateral incision may be required.

2.Terrible triad injury

In the terrible triad injury the elbow dislocates posterolaterally with an associated fracture of the radial head/neck with a fracture of the coronoid. There is usually an injury to the LCL.
Occasionally these injuries can be treated nonoperatively by close reduction of the elbow and immobilization in 90 degree of flexion for 7-10 days. Nonoperative treatment is carried out when the ulnohumeral and radiocapitellar joints can be concentrically reduced, the radial head fracture is minimally displaced and does not prevent rotation and the coronoid fracture is small. Post reduction the elbow must be sufficiently stable to allow early elbow mobilization.

Surgical treatment is indicated for dislocations with an unstable radial head fracture and a type III coronoid fracture (fractures involving more than 50% of the height).
Stability is restored from inside out and lateral to medial [6,7].
The coronoid is stabilized first with internal fixation or anterior capsular repair, followed by internal fixation or replacement of the radial head. Lateral and medial instability is restored by repair of  LCL and MCL [6,7].

3.Anteromedial coronoid fractures

About 58% (26-82%) of the anteromedial facet of the coronoid is unsupported by the proximal ulnar metaphysis and diaphysis and this makes it vulnerable to injury [8]. Anteromedial coronoid fractures are usually associated with disruption of LCL while the radial head and MCL remain intact [9].

O’Driscoll has classified coronoid fracture into 3 types:

• Fractures of the tip of coronoid
• Anteromedial fractures
• Fractures of the base (body) of the coronoid

The anteromedial fractures are further subdivided into subtype 1 (rim), subtype 2 (rim and tip), and subtype 3 (rim and sublime tubercle) [2].

X rays of the elbow can show these fractures but a CT scan is more useful in the identification of these fractures.

Most of the patients with this type of injury will require surgical fixation
because the elbow lacks stability as a result of the LCL injury and the loss of the medial buttressing effect of coronoid. Subtype 1 fractures can be treated with  LCL repair alone, and subtypes 2 and 3 fractures can be treated with internal fixation using cannulated screws, tension band, or a buttress plate.

Coronoid fractures which are minimally displaced (≤5 mm), or undisplaced with a concentric elbow joint, and a stable range of motion to a minimum of 30° of extension can be treated nonoperatively [10].

Simple elbow dislocation

In patients with simple elbow dislocation a closed reduction is carried out and the elbow is splinted in 90 degrees of flexion for 5-10 days. After about a week mobilization of the elbow is started. An extension block brace is used for 3-4 weeks. Recurrent instability after simple dislocations is rare (<1-2% of dislocations).

Complications of elbow dislocation

Some of the complications of elbow include:

• Failure of internal fixation --Most often seen after repair of radial neck fractures. Poor vascularity can lead to osteonecrosis and nonunion.
• Loss of terminal extension of the elbow--This is a common complication after closed treatment of a simple elbow dislocation. This can be prevent by early, active mobilization of the elbow.
• Varus Posteromedial instability--Injury to the LCL and fracture of the anteromedial facet of the coronoid can lead to varus posteromedial instability. This can be prevented by LCL repair and solid fixation of the anteromedial facet.
• Neurovascular injuries--Open elbow dislocations can be associated (rarely) with brachial artery injuries. Median nerve injuries can be associated with brachial artery injuries. Ulna nerve injuries can result from stretch injuries. 
• Compartment syndrome
• Post traumatic osteoarthritis
• Due to chondral damage and residual instability
• Recurrent instability
• Heterotopic ossification
• Contracture/stiffness
• Post traumatic stiffness --Common after complex dislocation. Early mobilization useful for prevention

Outcome of treatment

Simple dislocations

Generally the outcome of treatment of simple elbow dislocations is good with residual stiffness as a possible complication [11,12]. Recurrent instability can be a concern in patients where early mobilization is opted for whereas in patients where the elbow is immobilized for longer periods, elbow stiffness and contractures may be a problem [13]. Recurrent instability is usually not a problem in patients with simple dislocation. The incidence is low at about 0.3% [14]. There is some low quality evidence to show that the outcome at 2 years follow up in terms of pain and range of movements, favours early mobilization [14].

When the outcome is measure by MEPI score, quick DASH score and weeks off work, functional treatment appears to show significantly better outcomes.
There is no difference between surgical treatment of the collateral ligaments and plaster immobilisation of the elbow joint. Overall, functional treatment with early mobilization appears to provide better movements, less pain, better functional scores, shorter disability and shorter treatment time as compared to plaster immobilisation [14].

The is scarcity of good quality evidence, in literature, on the outcome of treatment of simple dislocations of the elbow.

Complex dislocations

The outcome of treatment of complex dislocations of the elbow have traditionally been poor due to the complexity of injury. Long term complications with such injuries include stiffness, pain, arthritis, and joint instability [15].

Chen et al [16] performed a systematic review of the literature to evaluate the the functional outcome and complications associated with the treatment of terrible triad injuries of the elbow (TTIE).
The review included a total of 16 studies all of which were retrospective in design, involving more than 300 patients. The overall, functional outcomes were satisfactory, but complications (including both those requiring reoperation and those not requiring reoperation) were common. The functional outcomes, as determined by Mayo elbow performance, Broberg-Murray, and/or DASH scores were consistently satisfactory.

Though a high proportion of patients had satisfactory functional outcomes, many patients developed complications, which included ulnar neuropathy, elbow joint stiffness, heterotopic ossification, and arthrosis. About a third of the patients required reoperation due to complications such as instability and/or elbow stiffness.

Rodriguez-Martin et al [17] did a review where they analysed the outcome of treatment of 137 elbow triad injuries in five published studies. The average follow up of the patients was 31 months. The overall outcome was satisfactory with an average flexion arc of 111.4 degrees, and average flexion of 132.5 degrees with forearm rotation of 135.5 degrees. The average Mayo elbow performance score was 85.6 points, and Broberg-Morrey score was 85 points. Complications related to ulnar nerve symptoms, post-traumatic arthritis, elbow stiffness and heterotopic ossification were not uncommon. The symptoms of post traumatic arthritis were mild to moderate in most of the patients who developed joint degeneration. Repeat surgical procedures included capsular release, hardware removal and secondary ulnar nerve transposition [17].

Conclusion

The elbow joint consists of three articulation, the ulnohumeral joint, radiocapitellar joint and the proximal radioulnar joint. It is one of the most inherently stable articulations of the skeleton. Static stability is provided the bony counters, capsule and the ligaments. The muscles crossing the joint provide dynamic stability.

Dislocations can be classified anatomically or based on the complexity of the dislocation. Usually the dislocations are classified as simple or complex. Simple dislocations which are not associated with fractures are easy to treat and the outcome of treatment is generally good with minimal complications.

Treatment of complex dislocations which are associated with fractures is much more difficult. The outcome of treatment of complex dislocations is less predictable and is often associated with more complications.

References

1. Ring D, Jupiter JB, Sanders RW, Mast J, Simpson NS. Transolecranon fracture-dislocation of the elbow. J Orthop Trauma. 1997;11(8):545–50.
2. O’Driscoll SW, Jupiter JB, Cohen MS, Ring D, McKee MD. Difficult elbow fractures: pearls and pitfalls. Instr Course Lect. 2003;52:113–34.
3. Mortazavi SM, Asadollahi S, Tahririan MA. Functional outcome following treatment of transolecranon fracture-dislocation of the elbow. Injury. 2006;37(3):284–8.
4. Mouhsine E, Akiki A, Castagna A, Cikes A, Wettstein M, Borens O, et al. Transolecranon anterior fracture dislocation. J Shoulder Elbow Surg. 2007;16(3):352–7. 
5. Wyrick JD, Dailey SK, Gunzenhaeuser JM, Casstevens EC. Management of complex elbow dislocations: a mechanistic approach. J Am Acad Orthop Surg. 2015;23(5):297–306.
6. Mathew PK, Athwal GS, King GJ. Terrible triad injury of the elbow: current concepts. J Am Acad Orthop Surg. 2009;17(3):137–51.
7. McKee MD, Pugh DM, Wild LM, Schemitsch EH, King GJ. Standard surgical protocol to treat elbow dislocations with radial head and coronoid fractures. Surgical technique. J Bone Joint Surg Am. 2005;87(1):22–32.
8. Doornberg JN, De Jong IM, Lindenhovius AL, Ring D, et al. The anteromedial facet of the coronoid process of the ulna. J Shoulder Elbow Surg. 2007;16(5):667–70.
9. Doornberg JN, Ring DC. Fracture of the anteromedial facet of the coronoid process. J Bone Joint Surg Am. 2006;88(10):2216–24.
10. Chan K, King GJ, Faber KJ. Treatment of complex elbow fracture-dislocations. Curr Rev Musculoskelet Med. 2016;9(2):185-9.
11. Morrey BF, An KN. Articular and ligamentous contributions to the stability of the elbow joint. Am J Sports Med 1983; 11: 315-9.
12. de Haan J, Schep NWL, Zengerink I, van Buijtenen J, Tuinebreijer WE, den Hartog D. Dislocation of the elbow: a retrospective multicentre study of 86 patients. Open Orthop J 2010; 4: 76-9. 
13. Schippinger G, Seibert FJ, Steinböck J, Kucharczyk M. Management of simple elbow dislocations. Does the period of immobilization affect the eventual results? Langenbecks Arch Surg 1999; 384: 294-7.
14. de Haan J, Schep NW, Tuinebreijer WE, Patka P, den Hartog D. Simple elbow dislocations: a systematic review of the literature. Arch Orthop Trauma Surg. 2009;130(2):241-9.
15. Rockwood CA, Green DP (1996) Rockwood and Green's fractures in adults. Philadelphia: Lippincott-Raven.
16. Chen HW. Liu GD and Wu LJ (2014). Complications of treating terrible triad injury of the elbow: a systematic review. PloS one. 2014, 9(5), e97476. doi:10.1371/journal.pone.0097476.
17. Rodriguez-Martin J, Pretell-Mazzini J, Andres-Esteban EM, Larrainzar-Garijo R. Outcomes after terrible triads of the elbow treated with the current surgical protocols. A review. Int Orthop. 2010;35(6):851-60.

Drug -- Alcohol Interaction: Medical Myth or Fact

           Drug -- Alcohol Interaction: Medical Myth or Fact

                   

                                             Dr. KS Dhillon

The mythconception that alcohol should never be taken with antibiotics dates back to the 1950’s. According to Karl S. Kruszelnicki [1], the venereal disease (VD) clinics of the 1950s and 1960s gave advice that alcohol should absolutely not be consumed while a patient was taking penicillin although there was no chemical interaction between penicillin and alcohol. Apparently, the advice was given for moral reasons. The doctors of the day were concerned about alcohol reducing the inhibitions of those having VD, while under the influence of alcohol, and they getting "frisky" and passing on the infection to other people before the penicillin could cure the sexually transmitted disease.

Alcohol and drug interaction

Some medications can interact with alcohol and alter the metabolism or effects of alcohol and/or the medication. There are two types of medication--alcohol interactions:
Pharmacokinetic interactions where the alcohol interferes
with the metabolism of the medication. 
Pharmacodynamic interactions where the alcohol enhances the effects of the medication, particularly in the central nervous system (e.g., sedation).

1.Pharmacokinetic interactions

Pharmacokinetics studies of the movement of the drug into, within and out of the body which essentially means what the body does to the drug.
About 10% of the alcohol consumed undergoes first-pass metabolism in the stomach, intestines, and liver. The major enzyme involved in alcohol metabolism is alcohol dehydrogenase (ADH), which converts the alcohol into acetaldehyde (a toxic compound) which is subsequently metabolized by aldehyde dehydrogenase (ALDH) to acetate. After the first-pass metabolism, alcohol goes to various parts of the body where it exerts its effect. Alcohol then goes back to the liver for metabolism and elimination. Beside ADH, CYP450 enzymes, mainly CYP2E1 are also involved in the metabolization of alcohol in the liver [2,3,4]. Alcohol consumption can alter the pharmacokinetics of certain medications by altering gastric emptying, affecting their absorption and metabolism. Similarly, certain medications can alter the pharmacokinetics of alcohol by altering gastric emptying and inhibiting gastric alcohol dehydrogenase. 

2.Pharmacodynamic interactions

Pharmacodynamics studies the effect and mechanism of action of drugs on the body which essentially means what the drug does to the body. Besides the effect of the alcohol on the central nervous system which produces impairment of performance and behavior, alcohol may contribute to the disease state which is being treated. An example would be impairment of gluconeogenesis which can lead to hypoglycemia in diabetic patients who are on treatment with oral hypoglycemic agents. A combination of nonsteroidal anti-inflammatory drugs and alcohol consumption can increase the risk of gastrointestinal hemorrhage.
Pharmacodynamic interactions involving alcohol and medications can increase the risk of adverse drug events and also increase susceptibility to the medications’ effect.

Therapeutic drug classes and drug-alcohol interactions 

There is a dearth of reliable studies on drug-interaction. Most of the evidence for drug-alcohol interactions is based on case reports and not on clinical trials [5].
There are three common therapeutic classes of drugs which interact with alcohol and these include antibiotics, cardiovascular drugs, and analgesics.

Antibiotics

Concomitant use of alcohol and certain type of antibiotics can cause or exacerbate the adverse effects. Disulfiram-like reactions have been reported in patients who consume alcohol while on the following antibiotics [6]:

• Cefamandole (Mandol)
• Cefoperazone (Cefobid)
• Cefotetan (Cefotan)
• Ceftriaxone (extremely rare)
• Chloramphenicol 
• Griseofulvin 
• Isoniazid 
• Metronidazole (Flagyl)
• Nitrofurantoin 
• Sulfamethoxazole (Bactrim)
• Sulfisoxazole

The disulfiram-like effect with cephalosporins is mediated by the inhibition of ADH, which in turn irreversibly inhibits the oxidation of acetaldehyde. The elevated concentrations of acetaldehyde produces facial flushing, nausea, vomiting, headache, tachycardia, hypotension, or a combination of all these effects. 
Metronidazole is also a known cause of disulfiram-like reactions when coadministered with alcohol. This reaction may involve ADH inhibition in the gastrointestinal (GI) tract, instead of in the liver as previously believed [7].  Disulfiram-like reaction has also been reported with the combined use of sulfamethoxazole/trimethoprim and alcohol [8].
Antitubercular drug, Isoniazid, is metabolized more quickly in chronic heavy alcohol users, and this can reduce the effectiveness of the drug [9]. Furthermore, the alcohol-isoniazid combination has been associated with an increased risk of hepatotoxicity and of disulfiram-like reactions [10]. Rifampin (Rifadin) and pyrazinamide which are also used in the treatment of tuberculosis are also known to increase liver toxicity when consumed with alcohol.
Ketoconazole (Nizoral), when combined with alcohol, may increase the risk of liver toxicity and disulfiram-like reaction.
Combination of cyclosporine and alcohol may increase the risk of central nervous system toxicity with possible seizures.
It is therefore imperative to advise patients to avoid alcohol intake for several days after consumption of antibiotic regimens known to interact with alcohol.
There is no scientific evidence to show that moderate alcohol consumption interferes with antibiotic effectiveness.

Cardiovascular Medications

Some medications for hypertension and angina are known to interact with alcohol. Nitrates such as hydralazine and nitroglycerin when taken with alcohol can increase the risk of orthostatic hypotension which may put the patient at risk of falls [3,4]. The metabolism of propranolol can be increased with chronic alcohol consumption, thereby decreasing the effectiveness of this beta-adrenergic blocking agent. Verapamil delays the elimination of alcohol and this can prolong alcohol intoxication [3,11]. Alcohol intake can also aggravate hypertension or heart failure [12,13].

Anticoagulants

Alcohol can interact with warfarin leading to an increase or decrease in its anticoagulation effect. Acute alcohol consumption can decrease the metabolism of warfarin leading to increased risk of hemorrhage. Chronic alcohol consumption increases the metabolism of warfarin leading to a decrease in the drug’s effect which can increase the risk of clot formation [3,4]. The exact mechanism of this alcohol and warfarin interaction is not known. History of alcohol consumption must be obtained from patients who are taking warfarin so that a close monitoring of the international normalized ratio (INR) can be carried out [14,15].

Antidiabetics

Diabetic patients who consume alcohol run the risk of hypoglycemia because alcohol suppresses gluconeogenesis. The risk of hypoglycemia is further increased if the patient is taking insulin or oral hypoglycemics [3,4].
Sulfonylureas such as tolbutamide and chlorpropamide when taken with alcohol are known to increase the risk of hypoglycemia. Excessive consumption of alcohol also increases the risk of diabetic complications such as diabetic neuropathy and retinopathy [3]. Heavy consumption of alcohol along with metformin intake can increase the risk of lactic acidosis [4]. Chlorpropamide, when taken with alcohol, can produce disulfiram-like reactions.

Non-Narcotic Analgesics

Nonsteroidal Anti-inflammatory Drugs (NSAIDs) and Aspirin

Case-control studies show that the use of NSAID or aspirin along with alcohol can increase the risk for an upper GI bleed. There is a three to fivefold increase in the risk of GI ulceration or major GI bleed when NSAIDs are used along with alcohol [16]. Alcohol consumption can cause gastritis by increasing gastric secretion and irritating the gastric mucosa.

Acetaminophen

Acetaminophen is an over the counter drug which is often used for pain relief. It is also found in combination with several narcotic medications. There have been several case reports which have reported severe liver injury in patients who use therapeutic doses of acetaminophen in conjunction with chronic alcohol use. Well-designed clinical studies to verify the validity of this interaction are, however lacking [11].
Acetaminophen is mainly metabolized through glucuronidation or sulfation (90%-96%) and also via CYP2E1 (4%-10%). CYP2E1 is the same enzyme involved in alcohol metabolism. Metabolization of acetaminophen through CYP2E1 produces a hepatotoxic metabolite called  N-acetyl-para-benzoquinoneimine (NAPQI). NAPQI is rapidly deactivated through hepatic stores of glutathione. When acute overdoses of acetaminophen are consumed, the stores of glutathione become exhausted and NAPQI accumulation can lead to fulminant liver failure [11].
In patients taking therapeutic doses of acetaminophen (<4 g/day) there is no increased risk of hepatotoxicity in alcoholic patients unless there are other risk factors [17].
The FDA advises patients that they should consult their doctor before taking over the counter pain medication if they take more 3 alcoholic drinks a day [18].

Narcotic Analgesics 

Opioid analgesics such as methadone and codeine derivatives can act on the central nervous system (CNS) to produce sedation and respiratory depression [19]. Alcohol also acts on the brain and can increase the risk of sedation and CNS depression in patients taking opioid analgesics [19]. There is a lack of studies on the interaction of alcohol and narcotic analgesics.

Besides these three common therapeutic classes of drugs which can interact with alcohol, there are other classes of drugs which can also interact with alcohol.

Antidepressants

Tricyclic antidepressants (TCAs) such as amitriptyline, doxepin, maprotiline, and trimipramine cause sedation and alcohol can increase the sedation caused by TCAs’ through pharmacodynamic interaction. Alcohol also interferes with the metabolism of amitriptyline in the liver thereby increasing the levels of amitriptyline in the blood. Furthermore, alcohol induced liver disease also impairs amitriptyline breakdown leading to significantly increased levels of active medication in the body. These high levels of amitriptyline lead to convulsions and disturbances in heart rhythm.
Selective serotonin reuptake inhibitors (SSRIs) such as fluvoxamine, fluoxetine, paroxetine, and sertraline which are now widely used as antidepressants, however, have much less sedating effect and no serious interactions occur when these are consumed with moderate amounts of alcohol [20].
Monoamine oxidase (MAO) inhibitors such as phenelzine and tranylcypromine can induce severe high blood pressure if consumed with tyramine, a substance which is present in red wine. Hence people taking MAO inhibitors should be warned against consuming red wine.

Antihistamines

Antihistamines, which are often used for the treatment of allergies and colds, are known to cause drowsiness, sedation, and hypotension, especially in elderly patients [21]. Alcohol through pharmacodynamic interactions, can enhance the sedating effects of these antihistamines and increase the patients risk of falling and affect one's ability to drive and operate machinery. Patients taking antihistamines should be warned against
consuming alcohol.

Barbiturates and Benzodiazepines

Sedative-hypnotic agents such as barbiturates (phenobarbital) and benzodiazepines (Valium, Xanax, Ativan) have a sedative effect and when consumed with moderate amounts of alcohol, synergistic sedative effects occurs leading to substantial CNS impairment.
These sedative-hypnotic agents can impair memory, just as alcohol can. Combination of these drugs with alcohol can exacerbate this memory-impairing effect.
Alcohol can inhibit the breakdown of barbiturates in the liver thereby increasing the blood of phenobarbital.

Conclusion

Though the alcohol and medication interaction has been reasonably well studied in chronic heavy alcohol consumers, the effect of moderate alcohol consumption has not been studied as thoroughly. Risks of oversedation in patients combining benzodiazepines and alcohol and the effects of warfarin and alcohol combination can be clinically very significant. The commonly held belief that patients taking antibiotic should not take alcohol is a myth since only a small number of antibiotics produce disulfiram-like side effect. Generally, alcohol does not reduce the efficacy of antibiotics. It is imperative that people taking either prescription or OTC medications, read product warning labels to determine whether possible interactions exist. Similarly, doctors prescribing medications should be alert to the possible interaction between alcohol and the medications been prescribed.

References

1. Kruszelnicki KS. Alcohol and Antibiotics. News in Science, 2 June 2005 at http://www.abc.net.au/science/articles/2005/06/02/1380836.htm accessed on 12/11/18.
2. Jang GR, Harris RZ. Drug interactions involving ethanol and alcoholic beverages. Expert Opin Drug Metab Toxicol. 2007;3:719-731.
3. Moore AA, Whiteman EJ, Ward KT. Risks of combined alcohol/medication use in older adults. Am J Geriatr Pharmacother. 2007;5:64-74. 
4. Fraser AG. Pharmacokinetic interactions between alcohol and other drugs. Clin Pharmacokinet.
5. Noureldin M, Krause J, Jin L, Ng V, Tran M. Drug-Alcohol Interactions: A Review of Three Therapeutic Classes. US Pharm. 2010;35(11):29-40. 
6. Weathermon R, and Crabb DW. Alcohol and Medication Interactions. Alcohol Research & Health. 1999; 23 (1): 40-51.
7. Tillonen J, Vakevainen S, Salaspuro V, et al. Metronidazole increases intracolonic but not peripheral blood acetaldehyde in chronic ethanol-treated rats. Alcohol Clin Exp Res. 2000;24:570-575. 
8. Heelon MW, White M. Disulfiram-cotrimoxazole reaction. Pharmacotherapy. 1998;869-870.
9. Baciewicz AM, Self TH. Isoniazid interactions. South Med J. 1985;78:714-718.
10. Isoniazid package insert. Eatontown, NJ: West-Ward Pharmaceutical Corp; March 2008.
11. Jang GR, Harris RZ. Drug interactions involving ethanol and alcoholic beverages. Expert Opin Drug Metab Toxicol. 2007;3:719-731.
12. Zilkens RR, Burke V, Hodgson JM, et al. Red wine and beer elevate blood pressure in normotensive men. Hypertension. 2005;45:874-879. 
13. Bau PF, Bau CH, Naujorks AA, Rosito GA. Early and late effects of alcohol ingestion on blood pressure and endothelial function. Alcohol. 2005;37:53-58.
14. Hylek EM, Heiman H, Skates SJ, et al. Acetaminophen and other risk factors for excessive warfarin anticoagulation. JAMA. 1998;279:657-662. 
15. Havrda DE, Mai T, Chonlahan J. Enhanced antithrombotic effect of warfarin associated with low-dose alcohol consumption. Pharmacotherapy. 2005;25:303-307.
16. Pfau PR, Lichtenstein GR. NSAIDS and alcohol: never the twain shall mix? Am J Gastroenterol.
17. Kuffner EK, Green JL, Bogdan GM, et al. The effect of acetaminophen (four grams a day for three consecutive days) on hepatic tests in alcoholic patients—a multicenter randomized study. BMC Med.
18. FDA proposes alcohol warning for all OTC pain relievers. U.S. Department of Health and Human Services. November 14, 1997. http://archive.hhs.gov/news/. 
19. Drug interactions. Thomson Micromedex. Greenwood Village, CO. www.thomsonhc.com. 
20. Matilla M.J. Alcohol and drug interactions. Annals of Medicine 22:363–369, 1990.
21. Dufour MC, Archer L and Gordis E. Alcohol and the elderly. Clinical Geriatric Medicine 8:127–141, 1992.

Surgery for removal of metallic implants after fracture union: Is it necessary?

Surgery for removal of metallic implants after fracture union: Is it necessary?

                                   Dr. KS Dhillon, MBBS, FRCS, LLM

Introduction

After fracture union metallic implants used for stabilization of the fracture serves no purpose, hence in the past, it was advocated that all such implants should be removed. This was partly due to fears of corrosion associated with the commonly used stainless steel alloy implants. However, subsequent research found such claims to be unfounded.

There are two broad class of fixation devices used in orthopaedic surgery for fixation of limb bones. The first group consists of wire or pins and the other includes screws, plates, and nails. In the past, most of the implants were made of stainless steel alloy. Now, however, titanium alloy implants are becoming more popular.

There are no clear guidelines in the medical literature on indications for removal of metallic internal fixation devices after fracture union. Surgeons have been making decisions about implant removal arbitrarily because of lack of evidence-based literature to guide them. The general rule has been that all wires should be removed while plates, nails, and screws may or may not be removed depending on prevailing circumstances.

The presence of metallic implants in the body usually does not produce any symptoms. However, under some circumstance, they can be a source of pain and or limitation of movement of the joints. In such circumstances, doctors are often compelled to remove the implants although not all such patients will be relieved of their symptoms after the surgery. Surgery for removal implants is not as innocuous as we often assume.

This review will probe what the current state of knowledge is regarding the need for implant removal after fracture union.

Types of implants used for internal fixation of limb fractures

Kirschner wires

 These are non-malleable stainless steel wires which come in different sizes and usually vary in diameter from between 0.7 mm to about 1.6 mm and in lengths of between 4 to 12 inches. There are usually used to fix small bone fragments which are not suitable for fixation with other devices such as screws, plates, and nails. Wires are frequently used to fix peri-articular fractures in children and also in adults, where non-invasive close reduction of the fracture can be carried out and the fracture stabilized with percutaneous wires. The frequent sites where wires are used for fixation of fractures include the elbow, wrist, hand, fingers, foot and the toes.

Cerclage wires

 These are malleable stainless steel wires which can be twisted around bones and tied in a knot. They are often used to fix fractures of the patella and the olecranon. They can also be used at other sites to complement fixation of bones with other devices.

Plates, screws, and nails

Various designs of plates, screws, and nails are available for internal fixation of fractures. These are usually made of stainless steel alloy or titanium alloy. The stainless steel implants are iron-carbon alloys with some element of chromium, molybdenum, and manganese. The titanium implants are alloys of titanium, aluminum and niobium.

Indications of implant removal

There is usually no disagreement among surgeons that K-wires that are used for temporary fixation of fractures should be removed after fracture union. K-wires protruding under the skin can cause skin irritation and pain. They can also migrate, if they are not securely fixed across two bone cortices, causing damage to other body structures. Cerclage wires can also cause skin irritation and pain if they are not buried in deep tissues and under such circumstance they should be removed. Screws that are subcutaneous and causing skin irritation should be removed.

Deep implants such as plates, nails, and screws may or may not be symptomatic and the removal such implants has been an area of debate and controversy. The questions that arise are whether the implants will cause harm if left in the body and do they cause any functional disability.

Do implants cause harm?

Biocompatibility of metallic implants has always been a concern because of the release of biologically active small particles due to oxidation of metal and the possibility of toxicity of these particles to the human body. Stainless steel alloys do corrode in the body but the implants become covered with a layer of fibrous tissue often as thick as 2 mm depending on the amount of corrosive material released and the amount of movement between the implant and the surrounding tissues.

Titanium alloys, on the other hand, do not corrode but they release ions which diffuse into the surrounding tissues [1]. It was believed in the 1970’s and 1980’s that these corrosive materials and metallic ions may be carcinogenic and predispose patients to cancers. However, experimental studies have not revealed any association between metallic implants and the development of any cancers [2]. Presently the possibility of corrosion and cancer are no longer considered to be an indication for removal of the implant [2]. Allergic reactions to metals in the body are rare and data substantiating implant related allergic reactions is scarce [2].

Another reason why plate removal was recommended in the past was due to the widely held belief that the presence of plates on the bone leads to bone atrophy. However, more recent studies have shown that if the plates are left in the body long enough the density of the bone returns to normal. Rosson et al studied the bone density in patients with forearm fractures and found that the bone density returns to normal after 21 months [3]. Similarly, studies of the tibia after plate removal have failed to show significant bone atrophy [4]. Hence plates can be left in the body without fear of plates causing bone atrophy or stress protection.

Essentially implants, left in the body after fracture union, do not cause any bodily harm. The next question that needs to be answered is whether implant produces symptoms or functional disability.

Do implants cause symptoms and when should they be removed?

There are very few circumstances under which implants would definitely need to be removed (absolute indications). K-wires can migrate and cause harm to other body structures and K-wires that are under the skin can produce pain, hence removal of K-wire would be indicated. Screws that perforate the joint should be removed because they can damage the joint when the joint is mobilized. Cerclage wires whose sharp ends are not properly buried under deep tissues can protrude under the skin leading to pain. Such wires obviously need to be removed. Implants such as plates adjacent to joints which are imperfectly positioned can obstruct joint motion and it would be necessary to remove them to improve joint function. Implants that are loose can migrate or produce irritation of adjoining soft tissues and such implants would also need to be removed.

Indication for removal of implants under most other circumstances is controversial and debateable. Some patients complain of pain or discomfort in the limb even when the implants are securely fixed and well positioned. The cause of such pain remains unclear and it is difficult to determine whether the implant is the cause of the pain or it is due to the injury itself [2]. In patients with such pain, the results of implant removal are ‘unpredictable and depend on both the implant type and its anatomic location’ [5].

Minkowitz in a study of 57 patients who had implant removal because of complaints of pain found that only 53% of the patients had complete resolution of pain at one year follow up [6].

Brown et al studied 126 patients who complained of lateral ankle pain in the region of the implants. Only 50% (11 out of 22 patients) had improvement of pain after implant removal. The functional scores were no different in patients who had and did not have implant removal [7]. The unpredictability of outcome has to be kept in mind when removal of implants for pain is contemplated.

Removal of implants involves another surgery which is accompanied with the risk of anesthesia-related complications as well as complications associated with the surgery itself. Cost of the surgery and hospitalization, as well as time off work, has to be borne in mind. Not only is the results of the surgery unpredictable, but the surgery itself can also sometimes be difficult and frustrating, resulting in broken implants and retrieval instruments. Surgical complications include bleeding, wound infections, neurovascular injury, refractures, recurrence of deformity, incomplete removal of hardware and sometimes poor cosmetic results because incisions for removal of the minimally invasive plate and nails are not so ‘minimal’.

Sanderson et al in a study of 188 patients who had implant removal found an overall complication rate of 20% and for forearm implant removal the complication rate was 42%. The nerve injuries that occurred were all permanent and were produced by junior doctors [8].

Richards et al in a smaller series of 86 adult patients who had a routine removal of implants in both symptomatic and asymptomatic patients reported a much lower rate of complications (3%) which included a nerve injury, a refracture, and a hematoma. However, the authors recommended that it would be appropriate to leave asymptomatic implants in situ [9].

Removals of implants from forearm bones are associated with higher complication rates. Langkamer et al [10] reported a 40% complication rate after forearm implant removal. Chia et al [11] reported a 27% and Bednar et al [12] reported a 10% complication rate following removal implants from the forearm.

Brown et al reported a 19% rate of significant complications in patients who had implant removal. They also found that patients who did not have their implants removed had no ‘appreciable problems’ and the authors recommended that implants should only be removed if there is a clear indication for the removal [13].

Karladani et al in a study of 71 patients who had removal of the tibial nail for pain found that only 39 of the 71 patients had improvement of pain and they were not totally pain-free after the surgery. In 14 patients the pain was the same and in 18 patients the pain was worse after the removal of the nail. Four of the 6 patients who had a previous fasciotomy were unhappy with the outcome of the surgery to remove the nail. The authors concluded that the outcome of tibial nail removal for pain is poor and that the nail should not be removed unless there are convincing reasons to do so [14].

There have been concerns that athletes with implants in situ, who participate in contact sports run the risk of a refracture because the implant can act as a ‘stress riser’. Evans and Evans did a retrospective study of 15 elite rugby players who returned to competitive sports while having implants in situ. Two of the players’ sustained implant related complications and the other 13 continued playing for up to 6 years without any symptoms. One of the players needed removal of wires which produced pain after the tension band wiring for a fracture of the patella while another player sustained an undisplaced fracture of the radius and had to be treated in a cast for one month. The authors concluded that an early return to sports after fracture union is feasible and it is not necessary to remove the implants which would further delay return to sports [15].

Routine removal of implants in pediatric patients is a common practice among many orthopaedic surgeons [16]. Kahle in arguing against routine removal of implants in children did a retrospective survey of 138 patients who had removal of implants and found a complication rate of 13%. Seven percent of the patients had an incomplete removal of the implant and 1.4% had a refracture. The study showed no evidence to support the policy of routine removal of implants [17].

Davids et al in a retrospective survey of 801 children with 1223 implants removed over a 17 years period reported a 12.5% complication rate of which 6% were major and 6.5% minor complications [18].

There appears to be no compelling reason to remove implants in children as is the case in adults after fracture union when the patient is asymptomatic.

Routine implant removal after fracture surgery consumes a remarkable portion of resources allocated for elective orthopedic surgery and is a potentially reducible consumer of hospital resources in trauma units [19].

The medical literature provides only level IV or level V evidence regarding removal of implants after fracture healing. Vos DI et al [20] have carried out a prospective multicentre clinical cohort study to evaluate the outcome of implant removal after fracture healing. The study included 288 adult patients, 146 patients had removal of upper limb implants and 142 patients had removal of lower limb implants. Removal of implants of the clavicle, humerus, radius/ulna, tibia, and femur was included.
The most frequent indication for implant removal was pain (63%) and limitation of joint motion (56%).

For removal of femoral nails, an extension of the scar was necessary in 62% of the patients and for tibial nails in 35% of the patients. For removal of upper limb implants, the estimated blood loss was between 0 to 300 ml and the operating time varied between 3 mins to 90 mins. For removal of lower limb implants, the blood loss was between 0 ml to 500 ml and the operating time was between 13 mins to 120 mins.

Thirty percent of the patients had one or more surgery-related complications. The complications included post-operative bleeding (11%), refractures (1%), nerve injuries (6) other complication in 9% of the patients. The complication rate for upper limb procedures was 22% while for the lower limbs it was 37%.

The number of patients with pre-operative complaints who had complete follow up was 214 (88%) and 6 months after the surgery the number of patients with complaints was reduced to 49% (P<0.0005). Despite the significant number of patients (39%) who had improved after implant removal there were 8 patients, who had no pre-operative complaints but developed 25 new post-operative complaints such as paraesthesia, pain, loss of strength and limited joint motion.

There appears to be no indication for removal of implants in patients, both adults, and children when there are no symptoms. In symptomatic patients the outcome of implant removal is unpredictable and the patient should be advised accordingly if implant removable is contemplated. In about half the patients the symptoms will not improve after the surgery. Furthermore, the surgery is not innocuous and between 3% to over 40% of the patients can develop complications depending on the type and location of the implant.

Conclusion

In the past routine removal of implants after fracture union was a common practice. This was because the metallic implants used for stabilization of the fracture served no purpose after the fracture had united and there were fears of carcinogenic toxicity of ions release form oxidative corrosion of the stainless steel alloy implants that were commonly used for fracture stabilization. However subsequent research has found that such claims are unfounded. Fears of bone atrophy and stress shielding related to the implants have also been found to be unfounded provided the implants are left in situ long enough. Hence implants need not be removed for these reasons.

However, under some circumstances, there are definite indications for removal of implants. There is no controversy or debate about removing K-wires, Cerclage wires, implants that penetrate joints and those that are imperfectly position and obstruct joint motion.

 In most other situations the indications for removal of implants are relative. Even stable infected implants, before fracture union occurs, are often left in situ but infected implants after fracture union should be removed to control infection. Implants that are loose which can sometimes migrate and cause symptom may have to be removed.

Deep-seated stable implants are usually asymptomatic and most authors recommend that they should be left in situ even in children. Removal of implants is not as innocuous as is often believed. Besides anesthesia-related complications, there are surgery-related complications in 3% to over 40% of the patients. Implant removal is also associated with increased cost and time off work.

In many patients who complain of pain around the site of the implant, the actual cause of the pain is often not known and it could be due to the effect of the injury rather than the implant itself. In symptomatic patients removal of implants resolve the symptoms in only about 50 % of the patients.

Before undertaking surgery to remove implants, it is imperative that the surgeon informs the patient about complications that can arise from the surgery. The patient also needs to be made aware of the unpredictability of the outcome of such surgery. Most authors are of the opinion that asymptomatic implants should not be removed and in other situations, the implants should only be removed if there is a definite compelling reason to do so.

References

1. Gotman I. Characteristics of metals used in implants. J Endourol. 1997; 11(6):383-9.
2.  Vos DI, Verhofstad MHJ. Indications for implant removal after fracture healing: a review of the literature. European Journal of Trauma and Emergency Surgery. August 2013, Volume 39, Issue 4, pp 327-337.
3. Rosson JW, Petley GW, Shearer JB. Bone structure after removal of internal fixation plates. J Bone Joint Surg [Br] 1991 ; 73-B :65-7.
4. Terjesen T, Nordby A, Arnulf V. The extent of stress-protection after plate osteosynthesis in the human tibia. Clin Orthop 1986; 207: 108-12.
5. Busam ML, Esther RJ, Obremskey WT. Hardware removal: indications and expectations. J Am Acad Orthop Surg. 2006 Feb;14(2):113-20.
6. Minkowitz RB, Bhadsavle S, Walsh M, Egol KA Removal of painful orthopaedic implants after fracture union. J Bone Joint Surg Am. 2007 Sep; 89(9):1906-12.
7. Brown OL, Dirschl DR, Obremskey WT. Incidence of hardware-related pain and its effect on functional outcomes after open reduction and internal fixation of ankle fractures. J Orthop Trauma. 2001 May; 15(4):271-4.
8. Sanderson PL, Ryan W, Turner PG. Complications of metalwork removal. Injury. 1992; 23:29-30.
9. Richards RH, Palmer JD, Clarke NM. Observations on removal of metal implants. Injury. 1992; 23:25-8.
10.  Langkamer VG, Ackroyd CE. Removal of forearm plates. A review of the complications. JBJS (Br) 1990 Jul;72(4):601-4.
11.  Chia J, Soh CR, Wong HP, et al. Complications following metal removal: a follow-up of surgically treated forearm fractures. Singapore Med J. 1996 Jun;37(3):268-9.
12. Bednar DA, Grandwilewski W. Complications of forearm-plate removal. Can J Surg. 1993 Feb;36(1):16.
13. Brown RM, Wheelwright EF, Chalmers J. Removal of metal implants after fracture surgery: indications and complications. J R Coll Surg Edinb 1993;38:96-100.
14. Karladani AH, Ericsson PA, Granhed H, Karlsson L, Nyberg P. Tibial intramedullary nails -- should they be removed? A retrospective study of 71 patients. Acta Orthop. 2007 Oct;78(5):668-71.
15.  Evans NA, Evans RO- Playing with metal: fracture implants and contact sport. Br J Sports Med 1997;31:319-321.
16. Jamil W, Allami M, Choudhury MZ, Mann C, Bagga T, Roberts A. Do orthopaedic surgeons need a policy on the removal of metalwork? A descriptive national survey of practicing surgeons in the United Kingdom. Injury. 2008; 39:362-7.
17. Kahle WK. The case against routine metal removal.  J Pediatr Orthop. 1994 Mar-Apr;14(2):229-37.
18. Davids JR, Hydorn C, Dillingham C, Hardin JW, Pugh LI. Removal of deep extremity implants in children. J Bone Joint Surg Br. 2010 Jul;92(7):1006-12. 
19. Böstman O, Pihlajamäki H. Routine implant removal after fracture surgery: a potentially reducible consumer of hospital resources in trauma units. J Trauma. 1996 Nov; 41(5):846-9.
20. Vos DI, Verhofstad MHJ, Vroemen. JPAM, van Walsum ADP, Twigt BA, Mulder PGH, van der Graaf Y, van der Werken. Clinical outcome of implant removal after fracture healing. Results of a prospective multicentre clinical cohort study. Submitted for publication. file:///C:/Users/user/Downloads/vos.pdf.

Limb Length Inequality: A Much Discussed but Little Understood Medico-legal Quandary.

                                 Limb Length Inequality: 

        A Much Discussed but Little Understood Medico-legal Quandary.

                              Dr K.S. Dhillon. MBBS, FRCS, LLM.

Introduction

In law, a person who is injured, due to the fault of another individual, is entitled to compensation for personal injury. In the assessment of damages resulting from the injury and the assessment of award for damages, the legal fraternity depends on a medical report prepared by doctors. The medical report is required to categorise and identify the severity of the injuries as well as to know the effects/ disabilities produced by the injuries. There is a significant difference between a medical examination for treatment and that for legal purposes. An examination for treatment is aimed at treatment and that for legal purposes is to ascertain damages and compensation. In the assessment for damages the asymmetry of the lower limbs and shortening of the limb is taken into consideration for award of damages. It is a common belief that the human body is symmetrical and many doctors as well as many in the legal fraternity are unaware that the human body is inherently asymmetrical.

In a normal population 90% of individuals have a 5.2 mm lower limb length inequality. In 32% of the population there is limb length inequality of between 0.5 to 1.5 cm and in 4% there is inequality of more than 1.5 cm. Limb length discrepancy of less than 2 cm does not appear to affect function and there is no need to equalise the limb length discrepancy. For accurate assessment of limb length inequality radiological imaging is required. Clinical assessment of discrepancy is a good screening technique but it is not a good technique for accurate assessment of limb length. The medical examiner has to take these factors into consideration before reaching a conclusion that a fracture has produced anatomical shortening or asymmetry.

Definition of limb length Inequality

Limb length inequality (anisomelia) could be due to structural asymmetry or functional inequality. In patients with structural asymmetry there can be actual shortening or lengthening of the limb. The actual structural skeletal length discrepancy is between the head of the femur and the ankle mortise and it could be in the femur or the tibia (not fibula). The causes of this discrepancy are varied and can be congenital/inherited or acquired. Congenital/inherited disorders include growth inhibiting conditions such as hypoplastic femur, Perthes disease and fibrous dysplasia and growth stimulating conditions such as vascular malformations, neurofibromatosis and fibrous dysplasia besides other causes.

The acquired causes of limb length discrepancy include growth inhibiting factors such as childhood fractures, infections and neurological damage (Polio, Cerebral Palsy, Head injury) and growth stimulating factors such as childhood fractures, haemangiomas, A-V malformations and infections [1]. In skeletally mature patients a common cause of acquired limb length discrepancy is due to fractures involving the femur or the tibia. However in the majority of patients the cause of the limb length discrepancy is not known even when there is a discrepancy of greater than 1 cm and it could be due to normal asymmetrical bone growth [2]. Genetic and environmental factors appear to influence ontogenesis of a person leading to this asymmetry [3].

Functional adaptation occurs in individuals with structurally short lower limbs which include foot pronation on the longer side and pelvic tilt leading to lower anterior and posterior iliac spine on the shorter side [3].

Functional inequality results when limb length discrepancy occurs in the absence of actual bone shortening or lengthening. It could be due to flexion contractures of the hip or knee and varus or valgus deformity of the knee besides other causes such as pelvic torsion and scoliosis. In functional limb lower limb inequality the pelvis rotates. The foot on the apparently short limb rotates externally, the heel goes into valgus and the foot arch collapses. The posterior iliac spine is higher on the apparently short side and the anterior iliac spine is high on the apparently longer side [3].

Classification of limb length discrepancy

The magnitude of limb length discrepancy has been classified into mild, moderate and severe [4].

• Mild – difference of less than 3 cm
• Moderate – difference of between 3 and 6 cm
• Severe – difference of more than 6 cm

The definition of biomechanically significant limb length discrepancy remains controversial among orthopaedic surgeons. While most are convinced that moderate and severe discrepancies are associated with some structural and functional disturbances, mild (less than 3 cm) discrepancies have not been convincingly linked to any structural or functional disturbances [3].

Prevalence and clinical significance of structural/anatomic limb length inequality 

Limb length inequality (LLI) appears to be a universal phenomena. Knutson in a review of published data on limb length inequality obtained by accurate and reliable x-ray methods found a prevalence of anatomic inequality in 90% of the subjects studied and the mean magnitude of anatomic inequality was 5.2 mm (SD 4.1). He concluded that anatomic leg-length inequality does not appear to be clinically significant until it reaches a magnitude of 20 mm (3/4 inch) [5]. The shortening is more common on the right side but the difference between the two sides is not statistically significant [5].

The most common effect of this limb length inequality is pelvic torsion on weight bearing with the pelvis rotating anteriorly on the short side and posteriorly on the long side. The clinical impact of these dynamic changes remains unclear [6]. Pelvic torsion is the commonest method of compensation with up to 22 mm of inequality. With greater degree of inequality knee flexion on the longer side is used as a compensatory mechanism [7].

Hellsing AL did a prospective study of young men during their military service to assess back function and pain before and after basic military training. Around 600 men were examined three times over a period of four years. The study found LLI of 0.5-1.5 cm in 32% of the subjects and 4% had a difference of more than 1.5 cm. However there was no correlation between LLI and back pain or pain provocative tests [8].

Gross RH in 1983 studied the incidence and the degree of limb length discrepancy in 35 marathon runners, 70% of whom were in the 30s and 40s. His hypothesis was that the occurrence of asymmetry would be lower in marathon runners than in general population because of attrition of runners with LLI. However he found that 34 of 35 runners participating in the study had LLI when measured by radiography. Eighteen runners (51%) had LLI of less than 5mm, 10 (28.5%) had a LLI of 5 to 9 mm and 7 (20%) had LLI of 1cm or more. He concluded that ‘runners can function handsomely without equalization’ of asymmetry and ‘that discrepancies of 5 to 25 mm are not necessarily a functional detriment to marathon runners, and no consistent benefits could be attributed to the use of a lift’ [9].

Gross RH in 1978 in a ‘survey of 74 skeletally mature patients with LLI of 1.5 cm or more found that patients with less than 2.0 cm did not consider their short limb to be a problem in any way’. As the magnitude of discrepancy increased there were problems but there was no critical ‘cutoff’ point. Some subjects functioned well athletically with discrepancies of over 2.5 cm. He concluded that there is little indication for equalisation of discrepancies of less than 2cm and for larger discrepancies clinical judgement has to be used on an individual basis [10].

Kaufman et al studied gait asymmetry in patients with LLI. They investigated 20 subjects (mean age, 9.0 +/- 3.9 years) who had documented limb-length inequalities, to determine what magnitude of discrepancies resulted in gait abnormalities. The study involved dynamic gait analysis which showed that generally, a limb-length inequality > 2.0 cm (3.7%) resulted in gait asymmetry which was greater than that observed in the normal population. This gait asymmetry however varied for each individual [11]. Less than 2 cm inequality does not seem to affect the kinematics [12].

A discrepancy in length of 5.5 per cent or more is associated with an increased mechanical work by the longer extremity and a greater vertical displacement of the centre of body mass. Clinically, this degree of discrepancy can be compensated by the use of toe-walking strategy while lesser degree of discrepancy can be overcome by a combination of compensatory strategies which normalize the mechanical work performed by the lower extremities [13]. Equalizing limb length can improve the symmetry of gait [14].

Gait asymmetry can easily be studied in a laboratory but it is more difficult to analyse the effect of limb length inequality on the spine, hip and the knee. The correlation between LLI and back pain, scoliosis, as well as knee and hip arthrosis has not been well established. The most commonly cited paper according to Knutson [5] which claims that LLI contributes to low back pain is that by Friberg [15].

 Friberg studied 1,157 subjects from a military hospital who were exposed to extreme and repetitive loading military activities. Of these 1,157 subjects, 798 had chronic low back pain and the control was 359 with no low back pain. In 75.4% percent of those with back pain and 43.5% of controls there was a 5 mm or more of LLI. Friberg’s findings were questioned by other investigators [5] and Friberg defended his results and clarified that ‘LLI of less than 5mm has no relationship to lumbar scoliosis or back pain’ and ‘that even marked LLI per se neither produces LBP or contributes to its development if a person is not continually exposed to prolonged standing or gait, e.g., doing daily work, military training and sporting activities’ [16].

 Soukka et al in a survey of 257 statistically matched working aged male and females subjects found that there was no increased risk of back pain with a LLI of 10 to 20 mm and that there was no conclusive evidence of increased risk of back pain with LLI of more than 20mm [17].

Gibson et al studied patients who had acquired LLI after femoral fractures. The study included patients between the age of 15 and 21 years who had sustained a fracture of the femur. There were 40 patients who were examined 10 or more years after the fracture. There were 15 patients with LLI of 1.5 cm or more who were studied further. The average LLI was 3 cm and ranged between 1.5 to 5.5 cm. During the 10 years only one patient wore a shoe raise and that too for a short period. None of the patients complained of low back pain and none had significant back pain over the 10 years. The acquired LLI produced little permanent structural abnormality in the spine and there was no degenerative changes seen in the spine after 10 years of the acquired LLI in skeletally mature individuals [18].

White et al studied 200 patients who had LLI after total hip replacement. They studied the relationship of LLI and functional outcome using the Harris hip score and the SF-36 Health Survey and found that leg lengthening (up to 35 mm) or shortening (up to 21 mm) had no correlation with functional outcome or patient satisfaction at six months after the surgery [19].

The association of LLI and lower limb complaints remains speculative. Tjernstrom and Rehnberg in a survey of 85 patients who had limb lengthening for shortening of between 3 and 14 cm (average 6cm) found that lower limb symptoms were not common in these patients and the effect of limb lengthening on joint symptoms was minor [20].

 There appears to be no good evidence that LLI leads to osteoarthritis of the hip or the knee. In 2007 Golightly et al studied the relationship of LLI and osteoarthritis (OA) of the hip and knee. They examined the relationship of LLI with radiographic hip and knee OA in a community based sample. There were 926 subjects with knee OA and 796 with hip OA, and 210 (6.6%) had LLI of 2 cm or more. They found that there was a significant association between LLI and OA of the knee but there was no significant association with hip OA. The weakness of the study was that limb length measurements were clinical with a tape rather than the more reliable radiological method. Furthermore the authors admitted that LLI in patients with OA could have been contributed by the loss of joint space, disfigurement of the joint and malalignment of the joint that occurs with OA. In additions to these drawbacks, the joints were not examined for contractures which can contribute to LLI. The authors proposed that future research should study the relationship of LLI to OA of the hip and the knee [21].

Harvey et al [22] in a recent large prospective study used radiographs for LLI measurements. Their cohort consisted of 2964 individuals with the age between 50-79, who were with or at high risk for knee osteoarthritis. They found leg length inequality ≥1 cm in 14.5% of the study subjects at baseline. In individuals with LLI of ≥1 cm there was significantly higher  prevalence of radiographic and symptomatic osteoarthritis at baseline. The predicted incidence of symptomatic knee osteoarthritis 30-months later was also higher in those with LLI of ≥1 cm. The shorter limbs were also at higher risk of of x-ray progression.

The authors however admitted that there are several weaknesses in their study. The number of incident radiographic OA cases were too small to effectively evaluate the effects of LLI and the authors were, for the same reason, unable to clearly define the risk associated with longer limbs and whether shorter or longer limbs were at higher risk of OA.

The second limitation of the study was the short duration of exposure due to the design of the study. A 30 month follow up was too short to detect progression of the OA. Furthermore even radiographic measurement limb length can be imperfect due to variation of knee flexion on standing radiographs. Hence there remains the possibility of misclassification and bias.

Gofton and Trueman [23] however stated that the LLI produces a pelvic tilt which leads to greater heel impact and stress on joints of the longer limb as compared with the shorter limb.

Kaj Tallroth et al [24] found that hip or knee arthroplasty due to primary OA was done 3 times more often in the longer lower limb than in the shorter limb. There were however some limitations in this study. No knee radiographs were taken at baseline examination and neither was information about  comorbidities, height, sporting activities and possible hip and knee problems during follow up, available. The only information available was the number arthroplasties done for primary OA of the hip or knee. In this 29 year follow up there was a surprisingly low incidence of OA [24].

Despite several studies having been published in the literature, the association of LLI and lower limb complaints remains largely speculative in nature.

Reliability of limb length assessment

Assessment of LLI is essential for appropriate treatment to be instituted. The limb can be measured clinically or by radiographic imaging techniques. Before using a particular technique it would be necessary to know the reliability (inter and intra-observer variation) and the accuracy (variation from actual length) of the technique used.

Clinical technique

The most convenient and commonly used method in clinical practice is the tape measure technique (TMM). The length of the lower limb is measured with a tape from the anterior superior iliac spine (ASIS) at the hip to the distal tip of the medial malleolus (MM) at the ankle.

However this method of measurement has some potential for errors, some of which include:

• Limb Girth- The size of the limb may vary between the two sides due to muscle wasting or due to swelling and this can produce an error.
• Difficulty in identifying the bony landmarks- The ASIS is difficult to palpate in obese individuals and MM can be difficult to palpate when there is swelling of the ankle. These bony landmarks may be deformed due to previous trauma making identification of the landmarks difficult.
• Angular deformities- Presences of angular deformities at the thigh, knee, leg and the ankle can introduce errors. A varus deformity will produce shorting while a valgus deformity will produce lengthening of the limb.
• Shortening distal to the ankle- This method does not measure shortening distal to the ankle mortise and can introduce error when there is bone loss distal to the ankle.
• Contractures- Contracture at the hip and the knee as well as pelvic obliquity can produce functional shortening when there is no actual structural length discrepancy.[25]

The problem of shortening distal to the ankle can be overcome by using an indirect method of measuring LLI. Blocks of known height can be place on the short side to obtain a level pelvis. This method also allows the treatment of LLI in patients with functional shortening.

A Galeazzi test is a quick and simple test which can be used to know whether the shortening is in the femur or the tibia. It is done with the person lying supine with the hip and knee flexed and the asymmetry of the level of the knees will confirm there is shortening and also reveal whether it is in the femur or the tibia. This test also eliminates the error of measurements when flexion contractures of the hip or knee are present. The actual length of the femur can be measured from the ASIS to the medial joint line and that of the tibia from the medial joint line to the MM.

Beattie et al studied the validity of tape measure method (TMM) for determining LLI when compared with a scanogram. They found that TMM derived measurement is a valid indicator of LLI and that the validity of the measurement is improved by taking the mean of two separate tape measurements. However the tape measurements were less reliable in healthy subjects as compared to those with LLI [26].

 Cleveland et al on the other hand showed that there is a statistically significant and weak correlation between radiographic measurements and physical examination measurements [27]. With tape measurements the accuracy in identifying the bony landmarks in obese patients, discomfort of the patient during the examination, and correct positioning of the patient by the examiner must be kept in mind.

Harris et al studied LLI in 35 patients with femoral fracture who were treated with nailing. Of these 35 patients, 15 patients (43%) had a measurable LLD. They found that ‘there was a positive correlation between direct leg length measurement and the block test (P = 0.003), and between the block test and patient perception of limp and LLD’. However, ‘there was no correlation between CT scanogram and clinical measurement of leg length or between CT scanogram and patient perception of LLD or limp’ [28].

Imaging technique

Imaging techniques commonly used for assessment of LLI include plain radiography and computerised radiography.

Plain radiography

 Techniques for measurement of LLI by plain radiography include:

• Orthoroentgenogram- This old technique requires a long cassette (longer than the limb length) with three exposures, each centred on the hip, knee and the ankle [29].
• Plain radiographic scanogram- This technique uses 3 ordinary cassettes lined up under the limb with the patient supine and a radio-opaque ruler tape to the table. The cassette is moved with each exposure centred on the hip, knee and the ankle [30].
• Teleroentgenogram- This technique involves a full length standing AP of the lower limbs using a long cassette with the X-ray beam at about 6 feet. The technique is similar to an Orthoroentgenogram but here a single exposure is used instead of three exposures. This technique is however subjected to magnification error.
• Computed radiography- A full length radiograph of the lower limbs can be obtained by obtaining 3 images on a vertical cassette holder containing 3 cassettes. The images are then transferred to a computer where a good quality full image can be downloaded and measurements done [25].

The inter and intra-observer reliability may vary from about 1.5mm to about 5mm. Up to about 10mm would be a threshold for a meaningful difference when assessing differences in measurements by these techniques [25].

CT scanogram

A CT scan is a very useful tool for assessment of limb length discrepancy. A scout view of both femurs and tibias can be obtained and the length of the femurs and tibias can be easily measured. The length of the femur can be measured from the tip of the femoral head to the distal end of the medial femoral condyle and the length of the tibia from the upper end of the medial tibial condyle to the tibial plafond. This technique appears to be more accurate than the plain radiography techniques [25].

The use of a tape is a cheap, simple and non-invasive screening technique for LLI. However it has potential for errors such as difficulty in palpation of bony landmark, difference in size of the limbs, angular deformities of the limb and contractures of the hips and knees. For more accurate assessment of LLI, radiographic techniques need to be used. The type of radiographic technique used will depend on the resources available. CT scans appear to be more accurate and reliable and will require less training and supervision of radiographers when compared with plain radiographic techniques [25].

Limb shortening/lengthening- Legal perspective

There is a significant difference between a medical examination for treatment and that for legal purposes. An examination for treatment is aimed at treatment and that for legal purposes is to ascertain damages and compensation.

Limb length inequality in the upper limbs usually does not produce any major functional impairment unless it is substantial and produces a cosmetic disfigurement. For example most fractures of the clavicle are treated conservatively and some degree of shortening is inevitable because the bone heals with overlap of bone fragments. However this shortening does not produce significant functional disability [31]. However lower limb length inequality can produce functional disability depending on the severity of the discrepancy between the two limbs.

Clinical measurements for treatment of the patient are commonly taken from the ASIS to the medial malleolus, irrespective of whether the shortening is in the femur or the tibia, because the goal of treatment is to compensate for the shortening when necessary. However the aim of measurements for medico-legal purposes is to determine whether the fracture of the femur or the tibia has resulted in shortening. Hence for medico-legal reporting it will be logical, persuasive and sensible to measure the length of the femur and the tibia individually. It cannot be assumed that when there is limb shortening it is due to the fracture of the tibia or the femur, because there can be inherent asymmetry between the two sides or there can be functional shortening due to angular deformity at the knee (valgus/varus) and/or flexion contracture at the hip or the knee. The length of the femur can be easily measured from the ASIS to the medial knee joint line and that of the tibia from the medial knee joint line to the medial malleolus.

Clinical limb length measurements are prone to errors due to differences in limb girth, difficulty in palpation of bony landmarks, angular deformities of the limb and contractures of joints. For medico-legal reporting, if initial clinical screening shows the presence of LLI, it would be logical, persuasive and sensible to provide accurate radiological measurements, to prevent conflict and make assessment of award for damages simpler.

Conclusion

Anecdotal evidence suggests that limb length inequality is a much discussed but little understood entity in personal injury litigation. It is commonly believed by both the medical as well as the legal fraternity that the human body is symmetrical but clinical evidence shows that the human body is inherently asymmetrical. Ninety percent of individuals in the normal population have a 5.2 mm of lower limb length inequality. In 32% of the population there is LLI of 0.5 to 1.5 cm and 4% have inequality of more than 1.5 cm. Limb length inequality of below 2cm does not appear to affect function and there is no need for compensatory correction of the discrepancy. Some individuals function well athletically even with discrepancies of over 2.5 cm.

It is commonly believed that limb length inequality leads to back pain, scoliosis, as well as osteoarthritis of the hip and knee but clinical evidence does not support such beliefs. The association between LLI and back pain as well as osteoarthritis of the lower limb joints remains largely speculative.

Clinical measurements of lower limb length are commonly included in medico-legal reporting by orthopaedic surgeons. However such measurements are prone to error. The presence of LLI on clinical screening should be followed with accurate radiological measurements so that conflict can be avoided and the award for damages justified.

References

1. Coleman NW and Sponseller PD. Limb-length discrepancy: Etiology, effects and evaluation. Orthopedic hyperguide at http://www.ortho.hyperguides.com/index.php?option=com_content&view=article&id=1375:limb-length-discrepancy-etiology-effects-and-evaluation&catid=308:operating_techniques.
2. Veterans Affairs Canada. Discussion Paper- Leg length Inequality (LLI) – Entitlement Eligibility Guidelines. March 2014 at http://www.veterans.gc.ca/eng/services/after-injury/disability-benefits/benefits-determined/entitlement-eligibility-guidelines/legleng
3. McCaw MC and Bates BJ. Biomechanical implications of mild leg length inequality. Br J Sp Med 1991; 25(1):10-13.
4. Reid DC, Smith B. Leg length inequality: a review of etiology and management. Physio Can 1984; 36: 177-82.
5. Knutson GA. Anatomic and functional leg-length inequality: A review and recommendation for clinical decision-making. Part I, anatomic leg-length inequality: prevalence, magnitude, effects and clinical significance. Chiropractic & Osteopathy 2005, 13:11 doi: 10.1186/1746-1340-13-11.
6. Cummings G, Scholz JP and Barnes K. The effect of imposed leg length difference on pelvic bone symmetry. Spine 1993; 18(3):368-373.
7. Walsh M, Connolly P, Jenkinson A, O'Brien T: Leg length discrepancy – an experimental study of compensatory changes in three dimensions using gait analysis. Gait Posture 2000, 12(2):156-61.
8. Hellsing AL. Leg length inequality. A prospective study of young men during their military service. Ups J Med Sci. 1988; 93(3):245-53.
9. Gross RH. Leg length discrepancy in marathon runners. Am J Sports Med. 1983 May-Jun;11(3):121-4.
10. Gross RH. Leg length discrepancy: how much is too much? Orthopedics.  1978;1(4):307-10.
11. Kaufman KR, Miller LS, Sutherland DH. Gait asymmetry in patients with limb-length inequality. J Pediatr Orthop. 1996 Mar-Apr; 16(2):144-50.
12. Goel A, Loudon J, Nazare A, et al: Joint moments in minor limb length discrepancy: a pilot study. Am J Orthop 26:852-856, 1997.
13. Song KM, Halliday SE, Little DG: The effect of limb-length discrepancy on gait. J Bone Joint Surg Am 79:1690-1698, 1997.
14. Bhave A, Paley D, Herzenberg JE: Improvement in gait parameters after lengthening for the treatment of limb-length discrepancy. J Bone Joint Surg Am 81:529-534, 1999.
15. Friberg O. Clinical symptoms and biomechanics of lumbar spine and hip joint in leg length inequality. Spine. 1983;8:643–651.
16. Friberg O. Letter-to-the-editor. Spine. 1992;17:458–460.
17. Soukka A, Alaranta H, Tallroth K, Heliovaara M. Leg-length inequality in people of working age. The association between mild inequality and low-back pain is questionable. Spine. 1991;16:429–431.
18. Gibson PH, Papaioannou T, Kenwright J: The influence on the spine of leg-length discrepancy after femoral fracture. J Bone Joint Surg (Br) 1983, 65(5):584-7.
19. White TO, Dougall TW: Arthroplasty of the hip. Leg length is not important. J Bone Joint Surg (Br) 2002, 84-B:335-8.
20. Bjorn Tjernstrom and Lars Rehnberg. Back pain and arthralgia before and after leg Lengthening: 75 patients questioned after 6 (1-1 1) years. Acta Orthop Scand 1994; 65 (3): 328-332.
21.  Golightly YM, Allen KD, and Jordan JM. Relationship of Limb Length Inequality with Radiographic Knee and Hip Osteoarthritis. Osteoarthritis Cartilage. Jul 2007; 15(7): 824–829.
22. Harvey W F, Yang M, Cooke T D, Segal N A, Lane N, Lewis C E, Felson D T. Association of leg-length inequality with knee osteoarthritis: A cohort study. Ann Intern Med 2010; 152 (5): 287–95.
23. Gofton J P, Trueman G E. Studies in osteoarthritis of the hip, II: Osteoarthritis of the hip and leg-length disparity. Can Med Assoc J 1971; 104 (9): 791–9.
24. Kaj Tallroth, Leena Ristolainen & Mikko Manninen. Is a long leg a risk for hip or knee osteoarthritis? Acta Orthopaedica. 2017;88: 512-515.
25.  Sabharwal S, and Kumar A, “Methods for Assessing Leg Length Discrepancy” Clin. Orthop. Relat. Res, vol. 466, no. 12, pp. 2910-2922, Dec. 2008.
26. Beattie P, Isaacson K, Riddle DL, Rothstein JM. Validity of derived measurements of leg-length differences obtained by use of a tape measure. Phys Ther. 1990;70:150–157.
27. Cleveland RH, Kushner DC, Ogden MC, Herman TE, Kermond W, Correia JA. Determination of leg length discrepancy. A comparison of weight-bearing and supine imaging. Invest Radiol. 1988;23:301–304.
28. Harris I, Hatfield A, Walton J. Assessing leg length discrepancy after femoral fracture: clinical examination or computed tomography? ANZ J Surg. 2005;75:319–321.
29. Green WT, Wyatt GM, Anderson M. Orthoroentgenography as a method of measuring the bones of the lower extremities. J Bone Joint Surg Am. 1946;28:60–65.
30. Moseley CF. Leg length discrepancy. In: Morrissy RT, Weinstein SL, eds. Lovell and Winter’s Pediatric Orthopedics. Philadelphia, PA: Lippincott Williams & Wilkins; 2006;1213–1256.
31. Nordqvist A, Redlund-Johnell I, von Scheele A, Petersson CJ. Shortening of clavicle after fracture. Incidence and clinical significance, a 5-year follow-up of 85 patients. Acta Orthop Scand. 1997 Aug;68(4):349-51.

Ankle fractures and post-traumatic osteoarthritis

              Ankle fractures and post-traumatic osteoarthritis

                                 Dr KS Dhillon FRCS, LLM

Introduction

There are several classifications for ankle fractures but the two commonly used are that by Danis-Weber and that by Lauge-Hansen. Familiarity with these two classifications is essential because published studies on ankle fractures use one or both these classification. The functional outcome of ankle fractures is good though there is a tendency to believe otherwise because the ankle is a small joint and is frequently injured. Despite the fact that the ankle is subjected to very high forces per square centimeter the incidence of post-traumatic osteoarthritis (OA) is low and the incidence of endstage OA is lower. The need for reconstructive surgery for endstage OA is very low. Although ankle replacement has been touted as a viable option for endstage OA, credible evidence for it is lacking. Ankle arthrodesis remains the gold standard.

Classification

Ankle fractures are usually classified according to a classification by Danis-Weber or that by Lange-Hansen.

Danis-Weber classification (1)

This classification is easy to use and is based on the level of the distal fibular fracture in relation to the tibio-fibular syndesmosis.

Type A – Infra-syndesmotic. The fibular fracture is below the level of the syndesmosis. The fracture is transverse and is due to an avulsion (stage 1- stable). The fracture may be associated with a vertical or oblique fracture of the medial malleolus (stage 2- usually unstable). About 20 to 25% of ankle fractures belong to this group.

Type B – Trans-syndesmotic. The fibular fracture is at the joint level and extends proximally in an oblique manner. Further force will produce a posterior malleolar fracture and if the force continues a tear of the deltoid ligament or an avulsion fracture of the medial malleolus will result (unstable fracture). This is the most common ankle fracture and constitutes about 60% of ankle fractures.

Type C – Supra-syndesmotic. The fibular fracture is above the joint line. Disruption of the syndesmosis occurs and there is an associated tear of the deltoid or an avulsion of the medial malleolus. These fractures are unstable and may be associated with a posterior malleolar fracture. External rotation of a pronated foot causes such fractures. The Type C fractures are not so common and constitute about 20% of all ankle fractures.

Lauge-Hansen (L-H) classification (2)

The L-H classification is based on the mechanism of the injury and is divided into 4 types depending on the directional force applied to the ankle.

Supination- External Rotation (S-ER) – This the most common type of ankle fracture comprising 40 to 75% of the fractures. There are 4 sequential phases of the injury.

1. Failure of the anterior-inferior tibio-fibular ligament (AITFL)
2. A spiral oblique fracture of the fibula at or above the ankle mortise
3. Failure of the posterior-inferior tibio-fibular ligament (PITFL) or fracture of the posterior malleolus
4. Tension failure of the Deltoid ligament or avulsion fracture of the medial malleolus

Supination-Adduction (S-AD) – These comprise about 10-20% of the fractures. There are 2 sequential phases of this injury.

1. Low avulsion fracture of the lateral malleolus or lateral ligament injury
2. Vertical shear fracture of the medial malleolus

Pronation – Abduction (P-AB) – These comprise about 5-20% of the fractures and usually there is a failure of the syndesmosis. There are 3 sequential phases of injury.

1. Failure of the deltoid or a transverse avulsion fracture of the medial malleolus
2. Failure of the AITFL and the PITFL
3. Transverse fibular fracture at or above the ankle mortise with comminution of the lateral cortex of the fibular 

Pronation – External rotation (P-ER). These fractures constitute about 7-19% of the ankle fractures and syndesmosis failure is common. There are 4 phases of the injury.

1. Failure of the Deltoid ligament or a transverse avulsion fracture of the medial malleolus
2. Failure of the AITFL
3. Spiral oblique fracture of the fibula above the ankle mortise
4. Failure of the PITFL or fracture of the posterior malleolus

Functional outcome and post-traumatic osteoarthritis

Functional Outcome

The outcome of treatment of ankle fractures is good. Egol et al (3) studied 232 patients who had surgical treatment of ankle fractures. At 1 year complete follow-up data was available in 198 patients (85%). Eighty eight percent of the patients at one year follow-up had no pain or mild ankle pain and 90% of the patients had no limitation or some limitation only in recreational activities. Young age, male sex, absences of diabetes mellitus, and a lower ASA class were predictive factors of good functional recovery after surgical treatment of ankle fractures at 1 year follow-up.

Lindsjö U (4) in a prospective study of 321 patients, with fracture dislocations of the ankle who were treated surgically, found excellent to good results in 82%, acceptable in 8% and poor results in 10% of the patients at 2 to 6 years follow-up. He found that the decisive factors that influenced the clinical outcome were the type of fracture, the accuracy of the reduction, and the sex of the patient. The outcome was good when the reduction was exact, fixation rigid and joint exercises were started early with early weight bearing with a below knee walking support. Men generally tend to do better than females.

Bauer et al (5) studied the natural history of ankle fractures in 143 patients who were treated conservatively by the closed method. The average follow-up was 29 years. Twenty patients had a Weber A, 103 Weber B and 20 patients Weber C fracture of the ankle. According the L-H classification 100 patients had S-ER, 15 S-AD, 14 P-AB and 14 P-ER fracture of the ankle. Eight three percent of the patients were symptom free at follow-up after an average of 29 years and 16% had occasional ache in the ankle. Eight two percentage of the patients had no OA and six patients (4%) had moderate OA and 2 patients (1.3%) had severe OA. The two patients with severe OA had a severe form of ankle fracture (S-ER type IV). The authors concluded that it is not necessary to have a perfect reduction of the fractures to have a good functional outcome in the treatment of ankle fractures.

Donken et al (6) in a study of 276 patients with S-ER type II – IV fractures who were followed-up for 21 years showed excellent or good results in 92% of the patients.
Complications associated open reduction and internal fixation are low. Nelson et al (7) studied the California (USA) discharge database of patients who had internal fixation of ankle fracture between 1995 and 2005. They found low short term complication rates which included a pulmonary embolism (0.34%), mortality (1.07%), wound infection (1.44%), amputation (0.16%) and reoperation for internal fixation (0.8%). Open fractures, older age, diabetes mellitus and peripheral vascular disease were predictors of short term complications.

Medium term complications in patients who were followed-up for five years included a 0.96% of patients who needed an ankle fusion or an arthroplasty. The medium term complications were more common in patients who had trimalleolar and open fractures of the ankle.

Post-traumatic Osteoarthritis

The primary cause of OA of the hip and the knee is idiopathic but the primary cause of OA of the ankle is trauma. Saltzman et al (8)) studied 639 patients with grade 3 and grade 4 (K-L) OA of the ankle who presented to a tertiary medical centre. They found that 70% of the ankle OA was post-traumatic while 12% was due rheumatoid arthritis and 7% was idiopathic. They also found that 8% of the hip OA was post-traumatic and for the knee the post-traumatic OA was seen in 12.5% of the patients.

Valderrabano et al (9)) found that 78% of the endstage OA of the ankle was post-traumatic in patients presenting at a tertiary medical centre. In 13% it was secondary and in 9% it was idiopathic. The cause of the post-traumatic OA was malleolar fractures (39%), ligament lesions (16%), pilon tibial fractures (14%), tibial fractures AO type 42 (5%), talus fractures AO type 43 (2%) and combined severe fractures (2%).

 The authors of the study did admit that there were two drawbacks in study. The first being the fact that the study was retrospective in nature it cannot provide the true prevalence rate of post-traumatic OA. The second important drawback is the fact that the study does not provide the exact pathomechanism of the OA in each etiological subgroup. It is easy to fathom that fractures of the distal tibia extending into the ankle joint can lead to secondary OA but how the fracture of the tibial shaft lead to OA remains unclear. One possible mechanism is that the fractures of the tibial shaft that lead to OA of the ankle have associated concomitant injury to the ankle which is not recognised.

Stuermer and Stuermer (10) in a prospective study of 214 patients with tibial fractures found that 20.1% of the patients had associated injury of the ankle. Of the 214 patients with tibial fractures, 45 ankles in 43 patients were found to have associated ankle injury. There was distal fibular fractures in 14, Maisonneuve fractures in 13, isolated rupture of the syndesmosis in 3, fracture of the posterior malleolus in 8 and fractures of the medial malleolus in 7 of the cases. In 38 of the 43 patients the syndesmosis was ruptured and 88.4% of the tibial injuries were spiral fractures located in the distal third of the tibia. Tibial fractures which have a potential for ankle injuries are those caused by pronation-eversion trauma, spiral fractures of the distal tibia, and those that are associated with proximal fibula fractures and fractures of fractures of the tibia with an intact fibula.

The studies by Saltzman and Valderrabano however do not provide the true prevalence rate of post-traumatic OA of the ankle. Lindsjö U (4) did a prospective study involving 321 patients with fracture dislocation of the ankle who were treated surgically and followed up for 2 to 6 years. He found a posttraumatic OA in 14% of the patients.

Bauer et al (5) who followed-up, 143 patients with conservative treated fractures of the ankle, for an average of 29 years showed a very low incidence of post-traumatic OA. Despite imperfect reduction of the fractures the incidence of moderate OA was 6% and severe OA 1.3%. The most common ankle fracture (30% of all ankle fractures) is the S-EV stage II and in this group of fractures only 1 patient out of 48 patients (2%) had a probable grade 1 OA. Kristensen and Hansen (11) found no OA in 94 patients with S-EV stage II fractures treated conservatively and followed-up for 16 to 25 years.
The latency period for development of endstage OA of the ankle after a fracture is rather long.

 Horisberger et al (12) in a study of 141 patients with ankle fractures who presented with endstage OA of the ankle found that the latency period between a fracture and endstage OA was 20.9 years (1-52 years). In 52.2% of the patients the endstage OA was due to malleolar fractures.

The number of patients who require reconstructive procedure after treatment of ankle fractures with open reduction and internal fixation is very low. SooHoo et al in population based study of 57,183 patients who had an open reduction and internal fixation of malleolar fractures found that 0.96% of patients required an ankle arthrodesis or ankle replacement on intermediate term follow-up (5 years). The mortality rate was 1.07%, infection rate 1.44 and amputation rate was 0.16% (13).

According to Daniels and Thomas (14) the ankle joint which is small and commonly injured is subjected to the highest forces per square centimeter. Despite these high forces the incidence of symptomatic ankle arthritis is much lower than larger joint such as the knee and hip due to various mechanical, biochemical and anatomic peculiarities of the ankle. These peculiarities make the ankle resilient to the process of aging and trauma.

Treatment of ankle post-traumatic OA

Non-surgical treatment

The pain in patients with ankle OA as in other joints is episodic in the early stages and such patients can be treated with analgesic or NSAIDs. In the later stages shoe wear medication can be useful. Shoes with cushioned heel or a rocker bottom sole modification can help reduce stresses across the ankle joint during heel strike. Braces and ankle orthosis can be useful but are often poorly tolerated. Corticosteroid injections can be useful in selected patients (15).

Surgical Treatment

Ankle (Tibiotalar) fusion remains the gold standard for treatment of symptomatic endstage OA of the ankle (15). It gives excellent pain relief but the drawback of the procedure is the loss of motion and it can lead to accelerated development of OA of other joints of the foot including the talonavicular, calcaneocuboid, naviculocuneiform, 1st Tarsometatarsal and the 1st metatarsophalangeal joints. The ideal position for fusion of the ankle is (16):

1. 5% degree of valgus
2. Neutral ankle position (no plantar or dorsiflexion)
3. 5%-7 degrees of external rotation
4. Slightly posterior positioning of talus in relation to the tibial plafond

Complications can be frequent with open ankle arthrodesis. Morrey and Wiedeman (17) did a review of 60 patients who had ankle fusion for post-traumatic OA. They found that 58% of the patients had a fusion within the first year of the injury. The infection rate was 23%, non-union rate 23%, and poor alignment and early loss of position was seen in 15% of the patients’ with 7% developing a delayed union. Forty one of the 60 patients were followed up for an average of 7.5 years and 83% were satisfied with the procedure. Thirty of the 41 patients had no motion at the subtalar joint but there was a13 degrees of motion at the Chopart's joint at an average follow up of 7.5 years. With shoes the gait of the patients was nearly normal.

Morgan et al (18) followed up 101 patients with ankle arthrodesis for an average of 10 years (from 2 to 25 years). There was a much lower rate of pseudoarthrosis of 5% and it was only seen in patients with sensory loss. Good to excellent clinical results was seen in 90% of the patients.

Ankle joint replacement

Ankle joint replacement has been touted as a viable option for treatment of end stage ankle arthritis. However the earlier reports showed failure rates as high as 72% (19).More recent studies have reported an 89% survivorship at 10 years but the quality of evidence in support of ankle replacement is weak and fraught with bias. High quality randomised control trials comparing ankle replacement with other forms of treatment for ankle arthritis is lacking (20).

It well established that the long term outcome of hip and knee replacements is excellent. However the clinical outcome of ankle joint replacement is not better that arthrodesis of the ankle as many would expect. Daniels et al (21) published a level II therapeutic study in which they studied the patient reported clinical outcomes in patients with ankle reconstruction. The data was obtained from the Canadian Foot and Ankle Society (COFAS) Prospective Ankle Reconstruction Database. The surgery was carried out by subspecialty trained orthopaedic surgeons.

The minimal follow-up was at least 4 years. There were 388 ankle reconstructions with 281 ankle replacements and 108 ankle arthrodesis. There was minimal difference between the Ankle Osteoarthritis Scale (AOS) and Short Form-36 scores between the two groups.

Seven percent of the arthrodesis and 17% of the ankle replacements underwent revisions. The major complication rate was 7% for arthrodesis and 19% for ankle replacements. The authors concluded that the intermediate term clinical outcomes of ankle replacement and ankle arthrodesis are comparable while the rates of revision and major complication were higher for ankle replacement.

Conclusion

Although the ankle is a small joint which is exposed to very high forces per square centimeter and is frequently injured the functional outcome of ankle fractures is good to excellent in 80 to 90% of the patients. Even with imperfect reduction of ankle fractures the incidence of post-traumatic OA is low and it ranges between 7 to 14% while the incidence of endstage OA is about 1.3%. In patients with S-EV stage II fractures (most common fractures) no OA develops even after 20 years. The latency period for endstage OA is about 20 years. The number of surgically treated patients with malleolar fractures who require ankle arthrodesis or fusion is low at about 0.96%.

Ankle replacement for treatment of endstage OA has not lived up to its expectation and credible evidence to support it as viable option for treatment of endstage ankle OA is lacking, unlike knee and hip replacements. Ankle arthrodesis remains the gold standard.

References

1. Bugler KE, White TO, Thordarson DB. Focus On Ankle Fractures. The Journal of Bone and Joint Surgery. 2012; 1- 4.
2. Clare MP. A Rational Approach to Ankle Fractures. Foot Ankle Clin N Am 13 (2008) 593–610.
3. Egol KA, Tejwani NC, Walsh MG, Capla EL, Koval KJ. Predictors of short-term functional outcome following ankle fracture surgery. J Bone Joint Surg [Am] 2006; 88-A: 974-9.
4. Lindsjö U. Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases. Clin Orthop Relat Res. 1985 Oct;(199):28-38.
5. Bauer M, Jonsson K, Nilsson B. Thirty-year follow-up of ankle fractures. Acta Orthop Scand. 1985;56:103–106.
6. Donken CC, Verhofstad MH, Edwards MJ, van Laarhoven CJ. Twenty-one-year follow-up of supination-external rotation type II-IV (OTA type B) ankle fractures: a retrospective cohort study. J Orthop Trauma. 2012 Aug;26(8):e108-14.
7. Nelson F. SooHoo, Krenek L, Eagan MJ, Gurbani B, Ko CY, Zingmond DS. Complication Rates Following Open Reduction and Internal Fixation of Ankle Fractures. J Bone Joint Surg Am, 2009 May 01; 91(5):1042-1049.
8.  Saltzman CL, MD, Salamon ML, Blanchard GM, Thomas Huff T, Haye A, Buckwalter JA and Amendola A. Epidemiology of Ankle Arthritis: Report of a Consecutive Series of 639 Patients from a Tertiary Orthopaedic Center. Iowa Orthop J. 2005; 25: 44–46.
9. Valderrabano V, Horisberger M, Russell I, Dougall H, Hintermann B. Etiology of Ankle Osteoarthritis. Clin Orthop Relat Res. Jul 2009; 467(7): 1800–1806.
10. Stuermer EK, Stuermer KM. Tibial shaft fracture and ankle joint injury. J Orthop Trauma. 2008 Feb;22(2):107-12
11. Kristensen KD, Hansen T. Closed treatment of ankle fractures: stage II supination-eversion fractures followed for 20 years. Acta Orthop Scand. 1985; 56:107–109.
12. Horisberger M1, Valderrabano V, Hintermann B. Posttraumatic ankle osteoarthritis after ankle-related fractures. J Orthop Trauma. 2009 Jan; 23(1):60-7.
13. SooHoo NF, Krenek L, Eagan MJ, Gurbani B, Ko CY, Zingmond DS. Complication rates following open reduction and internal fixation of ankle fractures. J Bone Joint Surg Am. 2009 May;91(5):1042-9. doi: 10.2106/JBJS.H.00653.
14. Daniels T and Thomas R. Etiology and biomechanics of ankle arthritis. Foot Ankle Clin. 2008 Sep; 13(3):341-52.
15. Ankle arthritis; OrthopaedicsOne Articles at http://www.aofas.org/education/orthopaedicarticles/ankle-arthritis.pdf.
16. Buck P, Morrey BF, Chao EY. The optimum position of arthrodesis of the ankle. A gait study of the knee and ankle. J Bone Joint Surg Am 1987;69(7):1052-62.
17. Morrey BF, Wiedeman GP. Complications and long-term results of ankle arthrodeses following trauma. J Bone Joint Surg Am. 1980 Jul;62(5):777-84.
18. Morgan CD, Henke JA, Bailey RW, Kaufer H. Long-term results of tibiotalar arthrodesis. J Bone Joint Surg Am. 1985 Apr;67(4):546-50.
19. Gougoulias, N., A. Khanna, and N. Maffulli.  How successful are current ankle  replacements?: a systematic review of the  literature. Clin Orthop Relat Res 2010; 468(1):199-208.
20. Zaidi R, Cro S, Gurusamy K, Siva N, Macgregor A, Henricon A, Goldberg A. The outcome of total ankle replacement: A systemic review and meta-analysis. Bone Joint J 2013;95-B:1500–7.
21. Daniels TR, Younger ASE, Penner M, Wing K, Dryden PJ, Wong H, Glazebrook M. Intermediate-Term Results of Total Ankle Replacement and Ankle Arthrodesis. J Bone Joint Surg Am, 2014 Jan 15;96(2):135-142