Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Ruptured Achilles tendons

Systematic review

There will be about 18 Achilles tendon ruptures for every 100,000 people. There is controversy between conservative treatment involving immobilisation, and surgical repair (which also involves immobilisation). All the textbooks you look at strongly recommend surgery. Is this right, and is it sufficient for creating a guideline or purchasing directive? A new meta-analysis [1] cautiously tells you that it is.

Systematic review

This McMaster review did some heroic searching for randomised trials comparing surgery with conservative treatment for Achilles tendon rupture. It also hand-searched journals, textbooks and abstracts. For inclusion studies had to involve patients with closed spontaneous rupture of the Achilles tendon, report re-rupture as an outcome, and be prospective randomised trials.

In addition textbooks of orthopaedics and narrative reviews were identified to assess whether they recommended (strongly or weakly) surgery or conservative management.


There were six trials with 448 patients. Different surgical techniques with 6-8 weeks of casting were compared with different conservative methods, all involving casting of various sorts for 6-8 weeks. Two trials used alternate assignments as the method of randomisation.

Surgery resulted in a significantly lower rate of re-rupture (3.1%) than did conservative management (13%). It also involved significantly more infection (4.7%) than conservative management (0%; Figure 1). There was no difference between the two techniques for return to normal function (about 70%) or spontaneous complaints (about 20%).

Figure 1: Re-rupture and infection after Achilles tendon surgery or conservative management

Recommendations about treatment of ruptured Achilles tendons were found in 22 review articles and textbooks. Of these 16 strongly favoured surgery, four were noncommittal, and two strongly favoured conservative therapy.


The best evidence we have is that for every 100 patients treated with surgery rather than conservative management, 10 fewer will have a re-rupture, but five will have an infected wound. The NNT was about 10 (95% confidence interval about 7 to 20) and the NNH was about 21 (13 to 58). There may be other adverse events of surgery not measured in these trials.

Tendon re-rupture is a serious setback, and antibiotics may cure wound infections. But with increasing rates of antibiotic resistance, wound infection may not be a trivial event. The confidence intervals of the benefit and harm overlap. There is certainly room for argument about which procedure is best.

And two of the six trials with 155 (35%) of the patients used randomisation methods which we know to be associated with some degree of bias. Judgement is more balanced than textbooks and reviews would lead us to believe. Of course, since the first randomised trials were published in 1981 techniques may have improved, and local skills may be different. But clear cut this decision is not, and we learn again that recommendations not based on evidence can lean towards the optimistic.

There is another point to be made. In Bandolier 102 , the way in which guidelines still tended to be relatively light on evidence was examined. Most guidelines rely on opinion, and do not quote systematic reviews or randomised trials. If guidelines were written on Achilles tendon treatment to reflect textbooks and narrative reviews, they would over-emphasise surgery compared with conservative management.

The real lesson is to follow the best evidence we have. Good systematic reviews should be the bedrock of guidelines, decisions, and future research.


  1. M Bhandari et al. Treatment of acute Achilles tendon ruptures. A systematic overview and metaanalysis. Clinical Orthopaedics and Related Research 2002 400: 190-200
previous or next story in this issue