Four-Strand Short Graft Failure Loads of Half–Peroneus Longus Tendon Grafts Compared With Full Semitendinosus Tendon Grafts

Methods

The research was performed in 2022 at the Queensland University of Technology, with ethics approval through Queensland University of Technology.

STTs and PLTs from both lower limbs of fresh-frozen cadavers were harvested by the 2 senior authors (D.F. and K.M.), who are both consultant orthopaedic surgeons. The tendons were labeled and stored at −80°C and then thawed for approximately 6 hours before preparation. The usable length of each tendon was measured before graft preparation. Grafts were prepared in standard Tape Locking Screw (TLS) 4-strand fashion (Figures 1 and 2). The TLS system for ACL reconstruction involves a short graft, small-diameter bone tunnels, and fixation by screwing the graft suspension strips from the outside in with a dedicated interference screw. ² Grafts were prepared with a 55 mm–long graft, looped around TLS polyethylene terephthalate tape at the graft ends, sutured at each end with a No. 1 Vicryl suture, and then tubularized with a running No. 3-0 Monocryl oversew. The tibial and femoral end diameters of each graft were then measured with the TLS graft sizing block, with the tibial end being the end first wrapped around the jig during graft preparation. The technique also involves preloading the graft to 500 N for 1 minute on the traction table of the TLS graft jig.

OPEN IN VIEWER

Figure 1.

(A) A peroneus longus tendon that has been split in half, and a Tape Locking Screw (TLS) graft preparation jig. (B) A tendon being prepared for a 4-strand TLS graft on the TLS jig.

OPEN IN VIEWER

Figure 2.

A prepared graft loaded on the pretensioning part of the jig.

An Instron 5567 machine was used to measure load to failure (Figure 3). Each graft was suspended in the Instron machine by looping and tying the TLS tapes at each end of the graft around the machine, and the grafts were preloaded to 10 N to tension the tapes, as otherwise this would have interfered with graft precycling. Each graft was then cycled 10 times up to 500 N, in order to align the tendon fibers inside the graft, ensure correct elastic behavior, and prevent stiffness from being previously frozen.

OPEN IN VIEWER

Figure 3.

(A) A prepared graft loaded on the Instron machine. (B) The Instron machine.

After precycling was complete, the machine returned to the rest position. Then graft extension was gradually increased at a rate of 20 mm/min, to simulate tendon rupture, with the load at the time of graft failure in newtons being measured (Figure 4). Data were recorded at a rate of 1000 Hz. Kurtosis and skewness measures were used to confirm that the data were normally distributed. Student t tests were then performed comparing the failure loads.

OPEN IN VIEWER

Figure 4.

A graft that completed precycling (cycles visible on the graph) and was being tested to failure.

Results

Seven cadavers were used in the study, with a mean age of 77 years, a mean height of 1.66 m, and a mean weight of 68.77 kg. Five of the cadavers were male and 2 were female. The first STT graft was used as an exploratory test and was excluded because it was used to calibrate the machine. One STT graft was excluded due to inadequate precycling and therefore was not tested according to the protocol for the study. Consequentially, 12 STT grafts were included, along with 14 half-PLT grafts used in total (Table 1 and Figure 5). The mean length of each graft before 4-strand preparation was 31.3 cm for the STTs and 27.15 cm for half-PLTs, and the mean tibial and femoral end diameters after graft preparation were 8.3 and 7.8 mm for the STT grafts and 7.9 and 7.3mm for the half-PLT grafts, respectively.

OPEN IN VIEWER

Table 1

Study Results

	Semitendinosus Tendon(12 Grafts)	Half–Peroneus Longus Tendon(14 Grafts)
Mean usable length before graft preparation, cm	31.3	27.15
Mean tibial end diameter of final graft, mm	8.3	7.9
Mean femoral end diameter of final graft, mm	7.8	7.3
Mean yield load, N	1291	964.5

OPEN IN VIEWER

Figure 5.

Yield loads for semitendinosus and half–peroneus longus tendon grafts.

The TLS technique involves preconditioning grafts to 500 N for 1 minute on the graft preparation table. Seven of 12 of the STT grafts and 8 of 14 of the half-PLT grafts were pretensioned according to the TLS technique. We did not pretension the other 5 STT or 6 half-PLT grafts because we wanted to see whether this affected the graft performance; however, we found no statistically significant differences in yield loads between the pretensioned and not pretensioned grafts. The mean load to failure of the preconditioned STT grafts was 1264.5 N compared with 1328 N for grafts that were not preconditioned (P = .58), and the mean load to failure for preconditioned half-PLT grafts was 959.4 N compared with 971.8 N for grafts not preconditioned (P = .91).

Overall, we found a statistically significant difference (P = .0001) between the mean load to failure for the STT graft group of 1291 N, compared with 964.5 N for the half-PLT grafts. All grafts were noted to fail at the tape-graft interface ends or via sutures giving way or cutting through, with no midsubstance tendon ruptures (Figure 6).

OPEN IN VIEWER

Figure 6.

A graft that failed, with the sutures cutting through the graft losing fixation, and the graft elongating.

Discussion

The primary finding of this study was that 4-strand half-PLT grafts demonstrated a significantly lower mean load to failure (964.5 N) compared with STT grafts (1291 N; P < .0001), showing that the hypothesis that half-PLT grafts would be comparable to STT grafts when used in a short 4-strand construct was incorrect. The TLS system uses a fixed-length construct that minimizes graft length in tunnels, typically 10 mm in the femur and 15 mm in the tibia, which has been established to be sufficient for graft healing. The graft preparation technique requires a minimum graft length of around 21 cm; however, this study found that the mean lengths of all tendons were more than adequate, with the mean STT measuring 31.3 cm and half-PLT being 27.15 cm before graft preparation.

A biomechanical study by Palmer et al ⁹ published in the American Journal of Sports Medicine in 2017 compared the properties of soft tissue allograft tendons and found a mean ultimate tensile force of 2294 N for PLTs and 1915 N for STTs when testing tendon strands in a CryoGrip fixture. Our method of suspending prepared grafts via their TLS tapes in the Instron machine, rather than securing tendon strand ends in a CryoGrip fixture, likely contributed to the lower failure loads observed in our study.

The loads experienced by the ACL during early rehabilitation are substantially lower than the failure loads observed in this study. Previous literature suggests that forces across the ACL during early range of motion and weightbearing exercises are typically in the range of 100 to 500 N. ⁸ Although the half-PLT grafts were significantly weaker than the STT grafts, they were still stronger than the expected loads to which a graft would be exposed in the early postoperative time period. Grafts typically become weaker during the early “ligamentization” period, and the potential exists for graft failure of the half-PLT grafts around 3 months postoperatively, as the rehabilitation loads are increased.^6,15

Graft failure in our study during testing on the Instron machine consistently occurred at the tape-graft interface, or due to suture cut-through, whereas the site of clinical failure is usually at the bone attachment sites, which is a potential confounder when translating our findings to clinical relevance.

Based on the results of our study, if a surgeon is considering using PLT for a 4-strand tendon graft, we recommend using the whole tendon. Current literature indicates no long-term functional morbidity from harvesting a full PLT, with American Orthopaedic Foot & Ankle Society (AOFAS) and Foot & Ankle Disability Index (FADI) scores being comparable to either preoperative scores, nonoperative sides, or hamstring harvest groups at between 6 months and 4 years follow-up in multiple studies.^5,7,10,11

The peroneus longus acts as a first ray depressor, whereas the peroneus brevis is an evertor of the foot. Patients are not able to isolate their peroneus longus or brevis muscles clinically. Many surgeons, including the senior author in this study (K.M.), when harvesting PLT, will tenodese the distal tendon end to the peroneus brevis to help maintain function.

Indications to consider using PLT graft can include patient preference (for reasons including a desire to avoid hamstring weakness and pain), previous hamstring harvest, and multiligament injuries requiring multiple grafts. In the practice of the senior author in this study, contraindications to using PLT for an ACL graft include preexisting ankle instability and occupations involving tiptoe activity, such as ballet and gymnastics.

Zhao et al ¹⁶ examined the strength of anterior half-PLTs harvested from 20 unpaired human cadaveric lower extremities by wrapping each end of the tendon around a hook attached to an Instron machine; those authors found a mean failure load of 322.35 N, which was comparable to the mean failure loads of STTs in their study. The second component of their article consisted of a retrospective study of 92 patients who underwent knee ligament reconstruction with an anterior half-PLT graft and were followed for >2 years; the investigators found no significant difference in pre- and postoperative AOFAS and FADI scores.

Rhatomy et al ¹² evaluated 31 patients who underwent ACL reconstruction with a full PLT graft; the investigators found no significant difference in ankle eversion strength or first ray plantarflexion strength between the donor ankle side and contralateral side 6 months after surgery.

Ertilav et al ⁴ examined 50 patients who underwent ACL reconstruction using a full PLT graft; the investigators found no significant differences in ankle eversion and plantarflexion strengths measured preoperatively and 4 years postoperatively and when compared with the contralateral ankle. As well, the investigators found no significant difference in preoperative and postoperative AOFAS scores. Ertilav et al noted that ankle range of motion angles of dorsiflexion, plantarflexion, inversion, and eversion were significantly lower in the postoperative period compared with the preoperative period and the contralateral side, with each direction reduced by a mean of 1° or 2°.

Agarwal et al ¹ compared 96 patients who underwent a hamstring ACL reconstruction versus 98 patients who had a full PLT ACL reconstruction; the investigators found no significant difference in the mean AOFAS score between the 2 groups but a statistically significant improvement in thigh muscle wasting in the PLT group 1 year postoperatively. They also found no significant differences between the preoperative, 6-month postoperative, and 1-year postoperative scores between the hamstring and peroneus longus groups for the International Knee Documentation Committee (IKDC) and Lysholm knee scores.

In a prospective cohort study, Saeed et al ¹³ followed 73 hamstring graft patients and 85 PLT graft patients for 24 months; those investigators found statistically significantly lower rates of donor site morbidity in the PLT group and no significant differences between the groups in the rate of graft rupture or in IKDC and Tegner-Lysholm knee scores at 24 months.

Our study has several limitations. The sample size was relatively small, which may limit the generalizability of the findings. The use of elderly cadaveric specimens does not accurately reflect the tissue properties of the younger population typically undergoing ACL reconstruction. Furthermore, this study focused on load to failure and did not assess other biomechanical properties such as stiffness or elongation with cyclical loading. Finally, as a time-zero biomechanical study, our study does not account for biological healing or in vivo conditions, which may significantly influence graft performance over time.

Conclusion

Four-strand half-PLT grafts had a statistically significantly lower load to failure compared with STT grafts in this study. Surgeons should use caution if planning on harvesting only the anterior half of the PLT or if splitting the tendon into 2 grafts.