Introduction

Hamstring (HS) grafts for anterior cruciate ligament (ACL) reconstruction have been associated with remarkably little donor site morbidity. HS grafts have long been recognized as the lower morbidity graft choice for anterior cruciate ligament reconstruction (ACLR) by comparison to bone-patellar tendon-bone (BTB). Their popularity took off when reports of high stability comparable to BPTB allowed them to be recommended for their efficacy as well as their safety. This chapter will discuss some of the pitfalls of HS use, although most do no really qualify as donor site morbidity, in the sense that, for example, patellar fracture after BPTB ACLR does.

Before discussing HS graft harvest morbidity, it is important to define what is meant when discussing HS grafts for ACLR. HS grafts are configured to form four-strand grafts in clinical use. This can be a four-strand semitendinosus (ST) graft or, much more commonly, a two-strand ST graft combined with a two-strand gracilis (Gr) graft. Although both are called HS grafts, only the ST is truly a HS tendon. The Gr is really an adductor tendon. Its primary motor function is adduction, and it is innervated by the obturator nerve, as are the other adductors. This is an important point because it helps explain the minimal morbidity attendant to harvest of these two tendons. The harvest represents the taking of only one HS tendon and one adductor tendon. Thus the morbidity and potential weakness are divided between two muscle groups. In both cases, the harvested tendon is the weakest mover within that muscle group. Studies have found no greater morbidity when Gr is harvested in addition to ST.

Hypesthesia

This is common after HS ACLR but diminishes gradually over time. It may result from damage to the infrapatellar branch of the saphenous nerve. This nerve branch is often sectioned in the course of making the usual anterior incision. It can seldom be seen, and results in permanent decreased sensation in a silver-dollar-sized area. This decreased sensation diminishes over time and is not of clinical significance. The use of an accessory posterior incision for harvest allows the anterior incision to be made quite small, which diminishes the chances of nerve branch transection. Greater numbness can be seen if the saphenous nerve or its sartorial branch is stretched or even cut. The nerve is generally not visualized during harvest but is closely aligned with the Gr tendon. The author has had a few cases where discomfort in addition to partial hypesthesia was present. In all cases, however, the discomfort diminished over a period of months, and in no case has it been a source of significant clinical morbidity, which is measured in the literature on the subject. A significant rate of hypesthesia occurs after HS harvest, but it is associated with any procedure that places an incision on the lower leg. As such, it is not really donor site morbidity. In addition, all reports in the literature describe it as diminishing over time and of no clinical significance to the patient.

Weakness

Studies have shown that knee flexor peak torque can be reliably restored to preharvest levels with diligent physical rehabilitation. While persisting minor HS weakness after HS harvest has been reported by some, other authors have found no measurable weakness after HS ACLR. Peak knee flexor torque is restored in the range where it is functionally important, in the 0–45 degree range , although not at 95 degrees of flexion. However, strong torque in this range is not used for any usual functional activity, and so this deficit is not clinically significant and is usually not noticed by patients. Weakness has never been reported to be a clinical or performance problem. Loss of internal rotation torque can also be measured, but again, no functional correlate to this has ever been found. It does not seem to matter whether an anterior or posterior harvest technique is used, nor does ipsilateral versus contralateral extremity harvest.

The potential for weakness is mitigated by diligent rehabilitation of the semimembranosus and biceps femoris to compensate for the harvested ST, and by the adductors magnus, longus, and brevis to compensate for the harvested Gr. Additionally, the ST tendon regrows within 1 year in most patients (see Chapter 30 ). Hip strength has not been found to decrease after HS ACLR.

Anterior Knee Pain

Anterior knee pain is found after all ACLR techniques. However, the incidence has been consistently found to be lower with HS grafts than patellar tendon grafts, and it is usually the same as with allograft use. This finding makes it likely that a certain baseline level of anterior knee pain is present in all ACL-injured patients, regardless of graft reconstruction type. Anterior knee pain probably represents true donor site morbidity only for the patellar tendon and, possibly, quadriceps tendon graft. One study found that BPTB resulted in shortening of the patellar tendon from scarring that is not present with HS grafts.

Infection

There has also been reported a very slight increase in infection rate (less than a 1% increment) compared with BPTB. However, this again is not a donor site problem but rather is associated with HS ACLR in general. A recent report has indicated a 0% HS ACLR infection rate when vancomycin-soaked sponges are used.

Cutting the Graft too Short to Use

Another potential problem with HS harvest, although it also is not donor site morbidity, is the possibility of cutting the graft too short to use. This could require the surgeon to use a different graft, although reports by Prodromos et al. and others have described techniques using a posterior incision that produce a nearly 100% success rate in obtaining a graft of adequate length. Indeed, a proper HS harvest produces significant redundant graft tissue if both the ST and Gr tendons are used. Thus even a suboptimal harvest should still generate enough tissue for successful reconstruction from the HS harvest.

True donor site morbidity, then, is limited to a few case reports, which will be reported in the following discussion.