Purpose of the study: Injury to the infrapatellar branches of the medial saphenous vein are incriminated in disorder of the anterior aspect of the knee after bone-tendon-bone ligamentoplasty procedures. We have demonstrated in an anatomic study the usefulness of a minimal two-way approach for harvesting the patellar transplant in order to preserve the nerve branches. The purpose of this clinical study was to evaluate the feasibility of this method and its impact on the sensitivity of the anterior aspect of the knee after ligamentoplasty in comparison with the usual harvesting technique.

Material and methods: This non-randomized prospective controlled contemporary study included 47 consecutive patients. The graft was harvested via two vertical incisions, one on the apex of the patella, the other on the eminence of the anterior tibial tuberosity. After harvesting the patellar splint, discision of the patellar tendon fibers was performed subcutaneously to the tibial tuberosity. Before removing the graft via the tibial incision with a forceps inserted via the inferior incision without injuring the peritendon. A tibial piece was then harvested. The ligamentoplasty was performed as usual using two anterolateral and anteromedial arthroscopic portals. The tibial tunnel was drilled first on the tibial tuberosity. These 47 knees were compared with 34 knees where the conventional approach was used (control group). We assess: harvesting time, width of the tendon transplant, quality of the graft, requirement to convert to conventional harvesting technique. Patients were reviewed at six weeks, three months and six months to assess anterior pain, dysesthesia, surface area of hypo or anesthesia and at six months kneeling problem.

Results: Conversion was not necessary for any of the knees. Mean harvesting time was 17 minutes (control group ten minutes). A good quality graft was obtained in all cases. Thirty-five patients were reviewed at six months. No sensorial disorders were noted in 18 patients. Sensorial disorders were noted in 17 patients (permanent hypoesthesia in the control group). None of the patients presented anesthesia. The mean surface area presenting a sensorial disorder was 13.6 cm² at six weeks (37.8 cm² in the control group) and 8.85 cm² at six months (23.4 cm² in the control group). Mean gain compared with the control group was 62%. There were two cases of anterior pain at six months and no case of dysesthesia. Sixteen patients could kneel normally (none in the control group); kneeling was not possible in one patient.

Discussion: The infrapatellar branches of the medial saphenous nerve are often injured when harvesting a bone-tendon-bone graft for ligamentoplasty. Anterior disorders would in part be correlated with the degree of sensorial impairment on the anterior aspect of the knee. The subcutaneous harvesting technique presented here with two minimal incisions appears to be an attractive alternative.

Conclusion: Our study confirmed the feasibility of this harvesting technique which significantly reduces the surface area of sensorial disorders and avoids most kneeling problems.