Dear Dr. Meals:

The Journal defines Level I evidence as a high-quality randomized controlled trial with statistically significant difference or no statistically significant difference but narrow confidence intervals. This study randomized sixty seven consecutive patients to either the stainless steel Teno Fix repair or a control suture repair. As with any randomized clinical trial, there were patients that did not receive the treatment selected by the randomization. In this trial all tendons were evaluated regardless of randomization for (1) Tendon dimensions sufficient to accept the device and (2) Transection location which would allow the distal anchor to be placed. As stated on pg 930 of the manuscript, of the 41 digits initially randomized to the Teno Fix, seven were reassigned to a control repair because of inadequate distal tendon for repair with the device. None of the changes in surgical repair were secondary to inadequate tendon size as the four tendons judged too small to accept the device anchors were all originally randomized to the control suture repair. The seven tendons which were switched to a suture repair at surgery were done solely because of an inadequate distal stump to allow for anchor placement. At the time of the study, techniques for intraosseous anchor placement were not available. Currently, in similar situations the anchor would be placed without difficulty with a transosseous technique.

Changes in repair technique for a distal injury pattern were not related to patient selection regarding anticipated compliance with therapy. In a subset analysis, there were no significant differences between the outcomes of the digits that were switched into the control group from the Teno Fix group. In addition, we have re-analyzed the data after omitting those digits which were changed during surgery to the control group. Our conclusions are the same with a dramatic decrease in the rupture rate in the Teno Fix group which remains the only statistical difference between the two groups.

We reaffirm that all patients were blinded to the repair technique as were the therapists and clinical evaluators . All patients were assigned an identification code for the study. The randomization code was not broken until the completion of the study. We are confident that this study qualifies as a high-quality randomized trial providing Level I evidence to the readership.

Regarding the management of FDS lacerations, we, along with other hand surgeons do not routinely repair FDS lacerations with concomitant FDP lacerations in order to minimize work of flexion underneath the pulley system. As stated on page 927 of the manuscript, the FDS stump was left as a gliding bed for the repair as well as to maintain the vascular supply to the FDP. The vinculae were preserved to protect the vascular input to the tendon if they were not already damaged by the original injury. Of note, FDS lacerations were not repaired in either group thereby not biasing the results toward either group. FDS lacerations were not left unrepaired in order to allow more room for the FDP tendon following repair. On the contrary, we feel that the single strand of 2-0 stainless steel suture that crosses the repair site is less bulky than the four stranded control suture repair. Regarding the effects of the repair on work of flexion, one of the authors the manuscript (E.D.) has recently performed a cadaveric work of flexion study comparing the Teno Fix repair to both 2 and 4-stranded suture repairs. (personal communication) The authors found no statistically significant differences in increase of work of flexion between the three repairs. Therefore, we encourage those authors who routinely repair FDS to continue to do so even when using the Teno Fix device.

Lastly we would like to address your observation regarding patient number 204, in which the device extruded secondary to infection. The patient reported that he was eighteen years old at the time screening and enrollment into the study. However, upon examination of established medical records after the patient had been enrolled, the patient was noted to be sixteen years old. We do not feel that the subsequent infection was related to the patient’s age but acknowledge that the patient should have been excluded from the study. We have reviewed the records of all remaining patients and find that they meet all inclusion criteria for this study.

We hope that this response has answered some of your questions and concerns regarding this clinical trial.

Brian W. Su, M.D.

Melvin P. Rosenwasser, M.D.

To the Editor:

With great interest, our local hand journal club recently discussed this article. Although we applaud the much needed and growing trend toward obtaining and reporting Level I evidence, we have several concerns on which we would appreciate having the authors comment.

We question whether this work truly qualifies as Level I evidence. Although the patients were randomized to suture or Teno Fix treatment groups before surgery, the surgeons changed certain patients from the experimental arm to the control arm of the study once they had exposed the lacerations. The manuscript states that only tendons large enough to contain the Teno Fix device were retained in that arm of the study. The opportunity for intra-operative alteration of the randomization plan for any reason clearly selects the more substantial tendons for the experimental repair and also introduces a bias to eliminate patients from the experimental group that may be perceived to be less compliant with the rehabilition protocol.

Additionally, it is not clear whether the patients were blinded to the type of repair they received. If not, there is once again the opportunity for bias on the part of the surgeon, therapist and patient regarding the type of the repair and the anticipated superior/inferior outcome.

On another note, among the 85 injured digits in both arms of the study, all had flexor digitorum superficialis lacerations in addition to flexor digitorum profundus lacerations. Only one FDS laceration was repaired. Acceptable current practice usually includes FDS repair along with FDP repair if for no other reason than to improve the vincular blood supply to the repaired deep tendon. We have concern about the Teno Fix device adding substantial bulk to the tendon repair and, of course, this effect would be minimized when the flexor digitorum superficialis is left unrepaired. Should JBJS's readers choose to use the Teno Fix, do the authors advise that the FDS not be repaired to allow for more room of the bulky FDP tendon?

Finally, we note in the Materials and Methods section that "all patients were at least eighteen years of age at the time of repair," yet in the Results section, one of the patients who developed an infection that led to extrusion of the Teno Fix device was "a sixteen-year-old boy."

We look forward to the authors' response to further our understanding of the proper application of this intriguing device for tendon repair.