| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Sign In to gain access to subscriptions and/or personal tools.
|
|
|
|||||||
Sign In to gain access to subscriptions and/or personal tools. |
|||||||||
Letter to the Editor |
Cincinnati, Ohio
Dear Editor:
We are writing this letter in reference to the article published by Thomas et al titled "Revision Anterior Cruciate Ligament Reconstruction Using a 2-Stage Technique With Bone Grafting of the Tibial Tunnel" (November 2005, pages 1701–1709). In this investigation, the authors followed 49 patients a mean of 6.2 years postoperatively. We would like to clarify a few issues the authors raised in the "Discussion" section regarding the results of our prior 2 publications10,11 on ACL revision reconstruction and potential reasons for some of the differences in findings between our results and those of the authors’ investigation.
We do congratulate the authors for recognizing the importance of correct anatomical placement of ACL grafts, which, in the types of cases they presented, required autogenous bone grafting first to achieve optimal placement. We discussed indications for staged tibial tunnel bone grafting in our prior study, as well as other options for handling widely misplaced or enlarged tunnels.10
First, it is important to note that Thomas et al excluded in their study 10 patients who had multiligament instabilities. This represents 17% of their total revision population of 59 patients. In addition, the authors’ instrumented test data showed that 9 of the 49 (18%) knees had only 3 to 4 mm of increased tibial displacement before the revision operation. The standard International Knee Documentation Committee (IKDC) definition of a deficient ACL according to instrumented laxity measurements is > 5 mm. We would speculate that these 9 knees had intact secondary restraints. In our prior investigations on ACL revision, all knees were included; 17 of 65 (26%) knees in one report11 and 17 of 55 (31%) knees in a second study10 had multiligament reconstructions. In addition, nearly all of the knees in our studies (116/120, 97%) had > 6 mm of increased displacement on instrumented laxity testing before the revision operation. The differences in populations and preoperative laxity could account for differences in the overall results.
Second, Thomas et al provided a failure rate of 24% from our publication on ACL bone–patellar tendon–bone autografts.10 Although this is correct for the entire population, we did stratify the group of 55 knees into those that had an ACL reconstruction only, those that required a staged high tibial osteotomy, and those that required a concurrent ligament reconstructive procedure. The group that required only an ACL reconstruction had a failure rate of 16%, which represented 5 of 32 knees. This failure rate was far greater than the 4% (13/312 knees) we reported after ACL bone–patellar tendon–bone primary reconstructions.12 We discussed the fact that ACL revision operations were performed in different patient subgroups and that the presence of other ligament deficiencies and the loss of secondary restraints to anterior tibial translation most likely affected the results in terms of a lower success rate. The mean tibial displacement in this group of 55 knees before surgery was 11 mm (range, 5–21 mm). This indicates that the majority of knees had indeed lost their secondary restraints before the revision reconstruction.
Third, Thomas et al alluded that the ACL failure rate in our allograft revision study of 33%11 was most likely explained by the use of 25 000 BY (2.5 Mrad) of gamma irradiation for graft sterilization. We disagree with this statement for several reasons. Our criteria for graft failure in that study were 2-fold: either > 5.5 mm of increased tibial displacement at follow-up or < 50% or < 3 mm of improvement in the knee arthrometer measurement compared to the preoperative value. These strict criteria are quite different from those used by Thomas et al in their study. In addition, there is no scientific evidence of an increased failure rate in allograft populations in whom low-dose irradiation is used when compared to those in whom no irradiation is used; the published failure rates have ranged from 5% to 19% in irradiated allografts7–9,13 and from 6% to 19% in fresh-frozen allografts.6,9,14–16
The American Association of Tissue Banks has advocated low-dose irradiation for many years, and we have used this method of sterilization in our allografts to improve protection against bacterial contamination. As of March 11, 2002, the Centers for Disease Control and Prevention had received 26 reports of bacterial infections from musculoskeletal allografts.1 Because the reporting of infections to the Centers for Disease Control and Prevention is voluntary, there could very well be other cases unknown as of present. The agency has recommended, "When possible, a method that can kill bacterial spores should be used to process tissue. Existing sterilization technologies used for tissue allografts, such as gamma irradiation or new technologies effective against bacterial spores, should be considered."1
Several animal studies have shown that no significant deleterious effects are induced by low-dose irradiation on graft healing properties. Curran et al2 reported that 2 Mrad of irradiation only produced 1 mm more of elongation of bone–patellar tendon–bone constructs under cyclic testing conditions compared to nonirradiated specimens. The authors stated that pretensioning and precycling allografts before insertion may eliminate the potential for some of this elongation in the early postoperative period. We routinely precycle our grafts before implantation. Other studies have shown that 2.0 Mrad of irradiation does not reduce graft stiffness, elongation, and strain.3,5 Therefore, we take exception to the statement of Thomas et al that "although the authors thought that the alteration in mechanical properties of the graft by irradiation was within acceptable limits, this hypothesis has not been proven, which may very well explain the relatively high incidence of allograft failure in the series."
Thomas et al mentioned the results of an investigation by Fox et al,4 who reported a failure rate of 6% of ACL revision reconstructions in which fresh-frozen allograft tissue had been used, with no secondary sterilization. The populations of these 2 studies were similar, as Fox et al also excluded patients with multiple ligament injuries. In addition, the preoperative knee arthrometer data showed that approximately 10% of their series had < 3 mm and 80% had only 3 to 5 mm of increased tibial displacement. Thus, although the knees in the study of Fox et al are comparable to those in the series of Thomas et al, they are not comparable to those of our allograft ACL revision study.
In our 1994 publication,11 we concluded that autogenous tissues should be used for ACL revision whenever possible, reserving allografts for cases in which an acceptable autograft cannot be harvested. The comment of Thomas et al that "Noyes et al now advocate that the allograft should not be considered as the first choice of graft revision surgery" therefore needs to be clarified, as our recommendation was published more than 10 years ago.
Although it is important for authors to continue to provide scientific data on ACL revision reconstructions to determine optimal graft constructs, we believe that the results of all revision knees should be reported. In our experience, approximately one third of these complex cases required multiligament reconstructions. To promote success rates in ACL revision reconstructions that are similar to those of primary ACL procedures, without including all revision cases, is not scientifically justified at present.
REFERENCES
Basingstoke, United Kingdom
This letter is in response to the letter by Noyes and Barber-Westin regarding the article "Revision Anterior Cruciate Ligament Reconstruction Using a 2-Stage Technique With Bone Grafting of the Tibial Tunnel" published in the November 2005 issue. We thank the authors for their interest as well as their comments.
The authors have raised 4 main questions, and this is an attempt to answer those queries.
The first concerns patients with multiligamentous instability. Over the study period, the senior author (N.P.T.) treated 18 cases of multiligamentous instability needing reconstruction using well-documented techniques. Of these 18 cases, the primary ACL reconstruction had failed in 10 owing to failure to recognize the associated laxity at the time of primary procedure. These patients needed revision ACL surgery and reconstruction of other ligaments (ACL + PCL, 3; ACL + posterolateral corner [PLC], 2; ACL + PCL + PLC, 3; ACL + PCL + PLC + lateral collateral ligament, 2). Our preference is to use autografts whenever possible. However, in these situations, autograft material is often scarce, and there is a lot of scarring and adhesions because of previous surgeries, and so we have used fresh-frozen allografts. Our preference is to use these grafts in extra-articular reconstructions. These patients need a different postoperative regimen, including support in a brace for a period of between 6 weeks and 3 months. All these patients were regularly followed in a dedicated research clinic, and the results were satisfactory. Their instability was abolished, and none of the patients have needed a further surgical intervention. In view of these obvious differences in the surgical procedure and postoperative regimen, we did not include this heterogeneous group of patients in the published study.
The second is about the clarification of results and failure rates after revision ACL reconstruction in the publications of Noyes and Barber-Westin. In general, revision ACL patients represent a mixed group (extent of meniscal and chondral pathology, synovitis, etc), even when other associated injuries (injury to PCL, lateral collateral ligament, PLC, and/or medial collateral ligament) are absent. This makes the comparison difficult, and therefore we decided not to include the patients with multiligamentous injury. We accept the fact that for patients with isolated revision ACL surgery, the failure rate in the hands of Noyes and Barber-Westin was 16% rather than 24%, as we seemed to have suggested, for this figure is accurate for their entire series. It is still higher than that in our published series.
The third concerns the use of allografts that have been sterilized with 2.5 Mrad of gamma rays. The authors have explained the rationale behind using the sterilization process, which in a way explains our reluctance for their routine use. In spite of taking all the necessary precautionary steps, it is quite possible that use of allografts can introduce a dormant infection into the host, which may manifest after many years, although the risk is likely to be small. The authors quote various articles comparing irradiated allografts with fresh-frozen allografts to prove the similarity between the 2 grafts. Although the results with both types of grafts are similar, the overall results (laxity measurements and revision rate) seem to be higher with the use of allografts as compared to autografts. Siebold et al2 compared the clinical outcome of 251 fresh-frozen patellar versus Achilles tendon allografts for primary ACL reconstruction and found comparable results in both groups, as stated by Noyes and Barber-Westin. Nevertheless, the authors stated that the total failure rate appears to be much higher compared with autogenous ACL reconstruction, indicating that the use of an allograft for routine, uncomplicated primary ACL reconstruction offers few advantages. Therefore, autograft tissue remains our graft of first choice for this procedure. We advise reserving allografts for revision procedures in which suitable autogenous tissues have been previously compromised, a contraindication for autogenous tissue harvest exists, or for multiple ligament surgery. This supports our preference for autografts.
The fourth question is about the definition of laxity. The criteria used in the literature for defining failure after revision surgery are variable, and this further makes comparison more difficult. The IKDC classifies knees that are within 2 mm of the normal contralateral knee by means of KT-1000 arthrometer or similar testing as being of "normal" stability. Knees that have greater than 5-mm difference are classified as having "abnormal" stability. The KT-1000 arthrometer applies a force of 134 N to assess knee laxity. For the past 15 years, we have used the Westminster cruciometer for laxity measurements. It is a validated tool, and during laxity measurement an 89-N force is applied.1 As the force applied is 89 N rather than 134 N, we have used comparatively lower figures to define patholaxity. We have also used the pivot-shift test as a criterion, as a positive pivot shift result (grade 2 or more) is suggestive of "functional" instability and has been shown to be a reliable indicator of symptomatic laxity. We used the criterion of grade 2 or more, as the IKDC defines the pivot-shift test result to be normal (when negative) or nearly normal for 1+ or glide, when compared to the contralateral normal knee. The same criteria were used to compare the results after primary ACL reconstruction. These criteria are stricter when compared to other series, including that of Noyes and Barber-Westin.
It is difficult to find an objective and agreed definition of "scientifically justified" as mentioned in the last paragraph of the letter by Noyes and Barber-Westin. We believe that this is a subjective view of the authors, and they are entitled to their opinion. In the interest of scientific clarity, we chose to report our clinical experience in this relatively homogeneous group with complete transparency. With the use of laxity measurements and clinical tests such as the pivot shift, we wanted to emphasize the important point that biological healing of the graft in the bone tunnel of a revision case could equal that of a primary reconstruction, provided that the wall of the tunnel comprised bone and not soft tissue from the previous procedure.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  J. H. Ahn, Y. S. Lee, and H. C. Ha Comparison of Revision Surgery With Primary Anterior Cruciate Ligament Reconstruction and Outcome of Revision Surgery Between Different Graft Materials Am. J. Sports Med., October 1, 2008; 36(10): 1889 - 1895. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
