Quadriceps grafts were bigger and stronger than bone-patellar-tendon-bone, and the harvester mattered

Design

controlled laboratory study with ultrasound measurement, cyclic loading and pull-to-failure testing

Population

30 grafts from 15 cadaveric knees: 15 bone-patellar-tendon-bone, 7 double-blade quadriceps, 8 cylindrical-harvester quadriceps

Primary outcome

graft cross-sectional area and biomechanical strength

Effect

cross-sectional area 0.44, 0.64 and 0.97 cm² respectively (all pairwise P<0.01); maximum load double-blade > cylindrical > bone-patellar-tendon-bone; stress and strain similar after normalisation for area

Thirty grafts were taken from 15 cadaveric knees: 15 bone-patellar-tendon-bone harvested with a 9 mm double-blade scalpel, 7 full-thickness quadriceps tendon grafts harvested with the same scalpel, and 8 quadriceps grafts taken with a proprietary 9 mm cylindrical harvester. Cross-sectional area was measured at three points by ultrasound by two raters, then each graft underwent cyclic loading and pull-to-failure testing.

Ultrasound measurement was reliable (intraclass correlation 0.70-0.87 for single raters). Cross-sectional area differed across all three: 0.44 cm² for bone-patellar-tendon-bone, 0.64 for the cylindrical-harvester quadriceps graft and 0.97 for the double-blade full-thickness quadriceps graft (all pairwise P<0.01). Maximum load and energy to maximum load followed the same order. The double-blade quadriceps grafts were also more variable in size than bone-patellar-tendon-bone, and the cylindrical-harvester grafts were the most variable in maximum load.

One result reframes all the others: after normalising for cross-sectional area, stress and strain were similar across graft types. The quadriceps tendon is not stronger tissue - it is more tissue. That matters clinically because it means graft size is the modifiable variable, the harvest technique determines it, and a surgeon can measure it before implanting: ultrasound cross-sectional area was reliable and correlated with strength. This is a cadaveric study of 30 grafts with no patient outcomes, and graft failure in vivo depends on fixation, tunnel position, biological incorporation and return-to-sport behaviour far more than on load to failure on a bench.

• Treat graft cross-sectional area, not graft type, as the variable you are controlling at harvest
• Expect a full-thickness double-blade quadriceps harvest to give the largest graft, and also the most size variability
• Consider measuring graft cross-sectional area before implantation rather than judging by eye
• Do not translate bench load-to-failure into clinical failure rates - the failure modes differ
• Remember donor-site morbidity differs between these harvests and was not assessed here

Why it matters

It shifts the graft choice argument from which tendon to how much of it you take.

Don't overread it

A cadaveric laboratory study of 30 grafts with no patient outcomes and no assessment of donor-site morbidity.

The statistics, in plain English

The equivalence of stress and strain after normalising for cross-sectional area is the most informative result here, because it explains the others: all three grafts behave like the same material, and the load differences follow from how much material each contains. With 7 and 8 grafts in the two quadriceps arms, the variability findings rest on very small groups and should be read as a signal rather than a measurement. Cadaveric tissue also fails differently from living tendon, which does not remodel on a testing rig.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.