Jerome A. Gilbert

Healing of experimental muscle strains and the effects of nonsteroidal antiinflammatory medication

The healing process of muscle strains and the effect of nonsteroidal antiinflammatory medication were studied using an experimental animal model. A standardized strain of the tibialis anterior muscle in adult male rats was produced by a controlled stretch of the muscle. Groups I and II underwent a unilateral strain of the tibialis anterior muscle and were immobilized in the postinjury period. The rats in Group II were fed piroxi cam in the postinjury period. Group III underwent a sham procedure and were also immobilized. At 0, 2, 4, and 11 days postinjury both injured and contralateral control muscles were evaluated by determining tensile strength characteristics and histologic appearance. Group I showed a significant drop in maximum failure load to 25.7% of the control leg at Day 2 with a gradual return to the level of Group III at Days 4 and 11 (40.9% and 50.1 %). Group II showed a similar drop but was still stronger than Group I at 2 days with 40.8% of the control leg and continued to drop until 4 days postinjury (33.7%). Histology showed a delay in inflammatory reaction and muscle regeneration in Group II. At 11 days both Groups I and II showed regenerated muscle fibers bridging the entire defect and an increase in endomyseal fibrosis. It is concluded that muscle strains continue to weaken in the early postinjury period. Non steroidal antiinflammatory medication, such as piroxi cam, might delay muscle regeneration.

The effect of a nonsteroidal antiinflammatory drug on the healing of ligaments

Because of the increasing number of ligament sprains being treated with nonsteroidal antiinflammatory drugs (NSAIDs), this study was undertaken to document the effects of one such drug on ligament healing in an experimental setting. Male Sprague-Dawley rats weigh ing between 400 and 500 g were used to evaluate the effect of the NSAID piroxicam on the healing of an experimental injury to the medial collateral ligament (MCL). The following factors were varied in the experi ments: dosage, days of treatment, and the day postin jury when treatment was begun. Piroxicam-treated rats were compared to placebo-treated rats in terms of the drugs effect on the mechanical strength of the healing ligament. The ligaments were mechanically tested in tension to failure at a constant deformation rate of 0.25 mm/sec on a materials testing machine. Administration of piroxicam on Days 1 to 6 postinjury resulted in a 42% increase in strength at Day 14 postinjury for the piroxicam-treated ligaments (P < 0.01) when compared with the placebo-treated controls. Neither doubling nor halving the standard piroxicam dose significantly al tered this increased healing strength. Biochemical analysis of collagen synthesis demonstrated a sugges tive, although not statistically significant, increase in collagen synthesis and collagen content in the piroxi cam-treated healing ligament. In separate experiments, piroxicam had no effect on the healed ligament at 21 days or on the strength of uninjured ligaments. In conclusion, piroxicam increased the early strength of healing ligaments in the rat when the drug was admin istered for short periods of time after injury. It did not affect the final strength when healing was complete, nor did it alter the strength of uninjured ligaments.

Displaced olecranon fractures in adults. Biomechanical analysis of fixation methods.

Four commonly used methods of internal fixation for displaced olecranon fractures were analyzed in a biomechanical model to determine strength of fixation. A transverse osteotomy in fresh cadaver specimens was internally fixed and was tested to failure by rapid loading. Screw plus wire in combination provided the greatest strength of fixation. Energy to failure averaged 31.0 Nm for cancellous screw, 32.0 Nm for figure-of-eight wire, 35.3 Nm for Arbeitsgemeinschaft fur Osteosynthesefragen (AO) tension band, and 39.4 Nm for screw-plus-wire combination. Comparison of screw and wire in combination to screw alone or to figure-of-eight wire demonstrated a significant difference (p less than 0.05). The difference between screw plus wire and AO tension-band fixation did not achieve statistical significance (0.2 greater than p greater than 0.1).

INTERNAL FIXATION OF FEMORAL NECK FRACTURES : A COMPARATIVE BIOMECHANICAL STUDY OF KNOWLES PINS AND 6.5-MM CANCELLOUS SCREWS

In a comparative biomechanical study of the stability of experimentally produced femoral neck fractures fixed with Knowles pins or with 6.5-mm cancellous screws, 12 matched pairs of anatomic specimen femurs were divided into four groups of standardized osteotomies. One member of each matched pair was fixed with three or four 6.5-mm cancellous screws, whereas the other member was fixed with the same number of Knowles pins. Mechanical testing was performed on an Instron testing machine. There was no statistically significant difference in maximum load to failure or in fixed stiffness between the two methods of fixation. When examining maximum load to failure, three of either device provided fixation equivalent to four. Although there was a difference in absolute fixed stiffness between specimens fixed with three Knowles pins compared to four, when stiffness was expressed as a percentage of intact stiffness, there was no difference between any fixation group. From an analysis of the observed modes of failure and from calculations of the flexural rigidity provided by the devices, it is proposed that the quality of the trabecular bone of the femoral neck is a major factor in the experimental fixation of femoral neck fractures.

Biomechanics of Ender rods, compression screw, and Zickel nail in the fixation of stable subtrochanteric femur osteotomies.

Compression screws, Ender rods, and Zickel nails were evaluated mechanically to determine their resulting stability in the fixation of experimental subtrochanteric femur osteotomies. Twenty-five pairs of human cadaver femurs were used. Experimental osteotomies were created and a random member of each pair was fixed with a compression screw (AMBI system, Richards Mfg. Co.). The other member was fixed with either Ender rods or a Zickel nail. The specimens were oriented in a direction to approximate the in vivo position of maximum resultant force on the femur and then subjected to a single progressive load to failure on a materials testing machine. The ultimate strength of fixation with the compression screw was ∼4.6 times greater than that with the Ender rods and 1.8 times greater than that with the Zickel nail. Ender rods and Zickel nails failed primarily in external rotation, whereas the compression screw failed by bone fracturing and varus bending.

Biomechanical studies of an anterior pelvic external fixation frame intended for control of vertical shear fractures

Biomechanical testing of an experimental pelvic external fixation frame incorporating a large tubular horizontal member was carried out on formalin-fixed cadaver pelves with simulated vertical shear injuries. The device was tested with 5 mm standard and 6 mm experimental pins, and was compared to a Pittsburgh triangular (PT) frame constructed of standard Hoffmann components. Tested by vertical loading, the experimental frame with 5 mm standard pins was 1.9 times as stiff as the PT frame, and with the 6 mm experimental pins it was 2.4 times as stiff as the PT frame. Clinical trials are needed to determine efficacy in vivo.

The effect of steam sterilization on plaster casting material.

Plaster-impregnated gauze was autoclaved as a preventative measure against infection. The mechanical properties and the hardening time of casts made with autoclaved plaster were investigated by measuring flexural yield stress and flexural modulus of elasticity. A cantilever bending test was used to determine the time required to harden. The average yield stress in bending was reduced by an average of 8% and the modulus of elasticity was unchanged by the steam sterilization. The time required for the plaster splint to harden was increased by a factor of 5.6 after steam sterilization. Although the reduction in the yield strength was not prohibitive, the increase in hardening time was unacceptable. The use of gas sterilization is recommended when a sterile cast is necessary because gas sterilization does not increase hardening time.

An evaluation of the bending stiffness of various tibial fixation methods.

The relative stiffness of human tibial fixation by plaster casts, intramedullary rods, external fixators, and dynamic compression plates was compared to the stiffness of the intact tibia. Four-point bending was performed on each specimen, first intact, and then after sequential instrumentation, thus allowing each specimen to serve as its own control. The mean stiffnesses obtained for the bone-device complexes, as a percentage of the intact tibial stiffness, were 113 +/- 9% for the delta frame fixator, 57 +/- 14% for the dynamic compression plate, 28 +/- 2% for the unilateral external fixator (UEF) with stainless steel sidebar, 18 +/- 1% for the UEF with aluminum sidebar, 7 +/- 2% for the flexible intramedullary (IM) rods, and 6 +/- 2% for the cast. Even relatively flexible unilateral external fixators are much stiffer than cast or IM rod fixation, methods that usually lead to satisfactory union by secondary healing. If secondary union with callus formation is desirable, modifications of the external fixator may be necessary to provide more flexible fixation. This study may be one of the first attempts to quantitate the stiffness of tibial fixation in a standard cast and compare it with other devices.

Experimental spiral fractures: an in vitro biomechanical comparison of lag-screw fixation to plate fixation

The rigidity and strength of fixation of experimental spiral fractures in canine tibias with triple lag-screw osteosynthesis were compared with fixation by six-hole dynamic compression plating (DCP). Lag-screw fixation was 71% to 77% as stiff in bending and 69% as stiff in torsion as plate fixation. When evaluated by torsion to failure, lag-screw specimens were 68% as strong as plate specimens. Although the difference in energy to failure in torsion was not statistically significant between the two methods, both fell considerably short of control values. The lag preparation was 9% as strong as intact controls, and the plate preparation was 21% as strong as intact controls. Long spiral fractures without comminution are amenable to triple lag-screw osteosynthesis with some loss of rigidity and strength when compared to single-plate fixation. This decrease in rigidity and strength may be outweighed by the avoidance of the stress protection phenomenon noted under rigid plates.

Reduction of stress shielding beneath a bone plate by use of a polymeric underplate: an experimental study in dogs

Plating of fractures results in osteopenia beneath the plate because of the stress-shielding effect. This study was performed to determine if a dynamic compression plate fixed to a bone with a low-density polyethylene underplate would loosen over time and with cyclic loading (due to creep and/or accumulated plastic deformation of the underplate) and thus decrease its stress shielding of the bone. Three groups of ten canine femora each were plated and tested in vitro. A nonosteotomized control group (without an underplate) and an experimental group (with an underplate) were cyclically loaded 10(6) times in compression under identical conditions. The third group of bones was osteotomized and plated with underplates and cyclically loaded 10(5) times to ensure that the construct would not fail before early fracture union could occur. The plate screws of the experimental group were significantly looser than those in the control group. This corresponded to a decrease in the underplate thickness that was more than sufficient to account for the looseness of the screws. Strain readings from the plate and bone were highly variable and not conclusive. Therefore, a transfer of load, reflected in the strain magnitude, from the plate to the bone during loosening of the screws could not be documented. None of the osteotomized bones in the third group had fixation failure. This experiment demonstrated that this plate-underplate construct provided strong initial fixation of an osteotomized bone in the laboratory setting without failure of fixation. This construct gradually loosened with time and cyclic loading in vitro and therefore should decrease the stress-shielding effect when applied in vivo.