Abstract
Background:
Many similarities exist between pyogenic flexor tenosynovitis and other closed-space infections such as septic arthritis. Previous studies have demonstrated that corticosteroids in conjunction with antibiotics considerably improve treatment outcomes in patients with septic arthritis. Using a chicken model, we investigated whether or not corticosteroids in combination with antibiotics and/or surgical drainage could minimize the loss of range of motion typically associated with pyogenic flexor tenosynovitis.
Methods:
We inoculated the flexor tendon sheath of the right long toe of broiler chickens with Staphylococcus aureus (American Type Culture Collection 29523 NA) (6 × 109 colony-forming units/mL) and twenty-four hours later administered one of six treatments to groups of fourteen or fifteen chickens. Treatment combinations included systemic or intrasynovial antibiotics, surgical drainage with catheter irrigation or no surgical drainage, and local corticosteroid injections or no corticosteroid injections. Measurements of active digital flexion at the proximal and middle interphalangeal joints were performed before inoculation and treatment and at seven, fourteen, and twenty-eight days after treatment. Flexion measurements were compared between groups as well as with similar measurements in the contralateral, uninfected, control long toe.
Results:
At twenty-eight days, two of three groups treated with locally administered corticosteroids and the group treated with intrasynovial antibiotics alone (without surgery) regained significantly more active flexion in comparison with chickens treated with systemic antibiotics and surgical drainage (the current standard of care). Pooled data revealed that the corticosteroid-treated groups regained significantly more active flexion at all post-treatment time points.
Conclusions:
Our data support the hypothesis that adding locally administered corticosteroids to the treatment regimen for pyogenic flexor tenosynovitis in a chicken model can significantly decrease loss of motion resulting from the infection. Furthermore, locally administered antibiotics may be effective for the treatment of pyogenic flexor tenosynovitis.
Clinical Relevance:
Additional clinical studies of the treatment of pyogenic flexor tenosynovitis with locally administered corticosteroids and/or antibiotics are warranted.
Pyogenic flexor tenosynovitis is a serious closed-space infection that can result in considerable morbidity if not promptly diagnosed and properly treated. Despite prompt treatment with appropriate antibiotics, surgical drainage, and irrigation, many patients develop sequelae, including stiffness or soft-tissue necrosis, which may ultimately require amputation1-3. Extensive adhesion formation may require flexor tenolysis1. The development of a treatment that could minimize the morbidity of pyogenic flexor tenosynovitis could benefit patients by helping them to regain more function of the affected digits.
Treatment for pyogenic flexor tenosynovitis is generally guided by the severity of the presentation, which may be gauged on the basis of the Michon classification scheme1. According to this system, pyogenic flexor tenosynovitis is classified as Stage I (increased fluid in the sheath, primarily a serous exudate), Stage II (cloudy/purulent fluid, granulomatous synovium), or Stage III (septic necrosis of the tendon, pulleys, or tendon sheath)1. Except for extremely early presentations of pyogenic flexor tenosynovitis, the standard of care for the treatment of this condition is surgical drainage, catheter irrigation, and appropriate systemic antibiotic coverage1,2. However, contention surrounds many issues in the treatment of pyogenic flexor tenosynovitis, including the timing of the initial treatment and whether or not open surgery or closed-catheter irrigation should be used for the initial treatment4-6, the method of irrigation used for the infection7-11, and the timing and method of antibiotic administration2,12,13. Because of the relative infrequency of this condition, clinical series evaluating different treatment modalities for these infections often have been small and rarely have exceeded twenty cases4,7.
Pyogenic flexor tenosynovitis is a closed-space infection. Other closed-space infections include septic arthritis, septic bursitis, infectious nephritis, and bacterial meningitis. Existing evidence indicates that these infections behave similarly and that novel treatments for one of these infections may be effective for another14-19.
Many similarities exist between pyogenic flexor tenosynovitis and other closed-space infections, especially septic arthritis. Both conditions involve bacterial invasion and replication in a nutrient-rich synovial fluid environment, and both involve considerable inflammatory-cell recruitment, leading to much of the morbidity associated with both conditions20. Previously, both animal and clinical studies have shown that the administration of corticosteroids in conjunction with appropriate antibiotics significantly improves clinical outcomes associated with the treatment of septic arthritis14-16. Of particular note is the double-blind placebo-controlled clinical study by Odio et al.16, which demonstrated that the addition of corticosteroids to the treatment regimen for septic arthritis in children was beneficial in terms of decreasing the duration of symptoms as well as the joint dysfunction resulting from the condition. However, we were unable to find any studies that have addressed the use of corticosteroids as an adjunct to antibiotics for the treatment of pyogenic flexor tenosynovitis.
Additionally, some studies, especially in the veterinary literature, have suggested that the treatment of pyogenic flexor tenosynovitis and septic arthritis should include local antibiotic administration21,22. One study showed that the administration of antibiotics into surgical wounds resulted in significantly decreased wound bacterial counts as compared with those following treatment only with systemic antibiotics23. Local administration also allows for the delivery of higher doses of antibiotics than could be tolerated systemically23. To our knowledge, no studies have evaluated the potential benefit of local antibiotics specifically for the treatment of pyogenic flexor tenosynovitis.
The purpose of the present experiment was to evaluate two important issues in the treatment of pyogenic flexor tenosynovitis. First, we wanted to elucidate what role, if any, corticosteroids should play in the treatment of pyogenic flexor tenosynovitis. We hypothesized that corticosteroids injected into the flexor tendon sheath in combination with antibiotics, with or without surgical drainage, would be as effective as, or more effective than, antibiotics and surgical drainage alone for the treatment of pyogenic flexor tenosynovitis, while minimizing long-term loss of range of motion. Our second goal was to establish the role that local antibiotic administration should play in the treatment of this condition. We hypothesized that a dose of locally administered antibiotics followed by systemic antibiotics would be more effective than systemic antibiotics alone for the treatment of pyogenic flexor tenosynovitis and would minimize the loss of range of motion resulting from the infection.
The research protocol was approved by the research institution's Institutional Animal Care and Use Committee.
Ninety adult (six-week-old) female broiler chickens were obtained from the Poultry Science Department of a veterinary school (North Carolina State University, Raleigh, North Carolina). These chickens had been bred for research and had been feed-restricted for the week prior to receipt in order to adjust them to the research environment of the study. The chickens were housed in a fenced pen with a floor of wood shavings, which were changed daily, and they were provided with as much as 114 g of chicken feed per day and water ad libitum.
Inoculation Procedure
Following a one-week acclimation period, the long toe of the right foot of all chickens was inoculated with 0.3 mL of gentamicin-sensitive Staphylococcus aureus (American Type Culture Collection [ATCC] 25923NA, Manassas, Virginia) (6 × 109 colony-forming units [CFU]/mL) via local injection into the tendon sheath with a 27-gauge tuberculin needle. The same investigator (R.W.D.) performed all inoculations. Staphylococcus aureus was chosen because it is the organism most commonly isolated from cases of pyogenic flexor tenosynovitis1,24. Injections were performed by inserting the needle into the pulp of the long toe between the metatarsophalangeal joint and the proximal interphalangeal joint until the needle struck bone. The needle was then slightly withdrawn to inject the inoculum of bacteria into the sheath overlying the proximal phalanx. The long toe of the left foot of each chicken did not receive an injection, in order to allow these toes to serve as uninfected, uninjected controls.
Experimental Groups
All chickens were examined with the help of the veterinary staff prior to the beginning of the study to ensure that none had obvious preexisting pathology involving the lower limbs or toes. Three of the ninety chickens were judged to be of poor health by the veterinary staff prior to beginning the study and were killed. The eighty-seven remaining chickens were randomly divided into six groups of fourteen or fifteen chickens on the basis of random number generation followed by a sorting procedure. The treatments of the chickens as well as the size of each experimental group are outlined in Table I. All chickens were treated with one of the six methods at twenty-four hours following inoculation, a time point that had been determined in pilot studies to result in Michon Stage-II pyogenic flexor tenosynovitis1.
Surgical Drainage and Other Treatments
Twenty-four hours following inoculation, four of the experimental groups (fifty-nine chickens) underwent surgical drainage of the infection. The same investigator (R.W.D.) performed all surgical procedures. Following adequate anesthesia with isoflurane (2.0% to 4.0% to effect), the right foot of the chicken was prepared with iodine solution and was draped in a sterile fashion. A small longitudinal skin incision (approximately 3 mm long) was made just distal to the metacarpophalangeal joint flexion crease of the long toe of the right foot with a #11 scalpel blade. With use of small scissors to dissect through the subcutaneous tissue, the tendon sheath was exposed and punctured. Another smaller longitudinal skin incision was made more distally on the long toe overlying the third phalanx, and a 22-gauge angiocatheter was then passed proximally through this incision into the tendon sheath. The sheath was then irrigated with 10 mL of normal saline solution. Both incisions were closed loosely with 6.0 monofilament polypropylene suture to allow further wound drainage following closure.
The groups receiving corticosteroids received a local injection of dexamethasone (Vedco, St. Joseph, Missouri) (0.06 mg/kg, 2 mg/mL) into the flexor tendon sheath immediately following surgical drainage and at two, four, seven, and fourteen days following the initial treatment. The groups receiving local antibiotics received a single local injection of gentamicin (Vedco) (5 mg/kg, 100 mg/mL) into the flexor tendon sheath immediately following surgical drainage or instead of surgical drainage. This was followed by systemic administration of gentamicin (5 mg/kg), delivered intramuscularly, twice daily for five days. Injections of both corticosteroids and gentamicin into the flexor tendon sheath were performed overlying the proximal phalanx of the long toe, between the metatarsophalangeal and proximal interphalangeal joints. The test groups that were treated with systemic antibiotics received a dose of gentamicin (5 mg/kg) intramuscularly immediately following surgical drainage, followed by systemic gentamicin (5 mg/kg) intramuscularly twice daily for five days following surgical drainage or experimental treatment. All chickens received butorphanol (1 to 4 mg/kg) intramuscularly for pain relief once daily for three consecutive days following treatment. Similar to other experimental interventions, the same investigator (R.W.D.) performed all injections of corticosteroids and gentamicin.
Assessment of Range of Motion
Immediately before inoculation, twenty-four hours after inoculation (but before treatment), and seven, fourteen, and twenty-eight days after treatment, active range of motion of the experimental long toes and the contralateral, control long toes was assessed by an investigator (R.W.D.) who was blinded to treatment group. This was done by measuring the flexion of the digit at the proximal and middle interphalangeal joints with a Rolyan Flexion/Hyperextension Finger Goniometer (Homecraft Rolyan, Huthwaite, United Kingdom) (Fig. 1). This was performed in a method similar to that described by Jaibaji et al.25. The finger goniometer was placed on the dorsum of the digit in an attempt to take each flexion measurement as accurately as possible and was performed by the same investigator (R.W.D.) throughout the experiment. Similar to the method employed by Jaibaji et al., maximum flexion of the digit was elicited by stroking the plantar surface of the foot, which causes reflexive, full flexion of all toes in these chickens25. Also, similar to the method described by Jaibaji et al., the interphalangeal flexion measurements obtained from the proximal and middle interphalangeal joints were then added together for data analysis25.
To assess a more clinically relevant measure of active range of motion, the chickens were perched on progressively smaller perches prior to inoculation and treatment and on Post-Treatment Days 7, 14, and 28 by an investigator (R.W.D.) who was blinded to treatment group. The perches had progressively smaller diameters of 2.8 cm for the largest-diameter perch, 2.2 cm for the medium-diameter perch, and 1.6 cm for the smallest-diameter perch, similar to a method that has been reported previously25,26. We elected to use the smallest-diameter perch possible after determining the perch diameter that all of the chickens in our cohort could easily grasp. This perch diameter corresponded with the smallest perch diameter reported in a previously reported assessment of chicken perching ability25. A chicken was judged to successfully perch on a given diameter only if the pulp of the distal portions of the long toes of both feet contacted the perch.
Statistical Methods
The number of chickens selected for each treatment group was based on a sample size calculation for paired samples with use of the following assumptions: alpha = 0.05, power = 0.8, delta = 15° (roughly the difference in flexion sums [proximal interphalangeal joint flexion + middle interphalangeal joint flexion] in chickens that could grasp the perch with the smallest diameter versus the perch with the second-smallest diameter, as determined in pilot studies). The standard deviation of the differences of the response of matched pairs in each group was 12.5°. With use of these assumptions, the minimum sample size was calculated to be n = 8. We decided to use groups of fifteen chickens to ensure that the study was sufficiently powered to detect a lack of difference between the active flexion measurements of the experimentally treated toe as compared with the contralateral, control toe.
Routine descriptive statistics including means, standard deviations, and 95% confidence limits were calculated for the active flexion measurements of both feet at each time point. Differences between experimental and control-toe active flexion measurements were determined with use of a paired one-way analysis of variance. Differences between the experimental and control toes in each group were compared with the differences in other experimental groups with use of analysis of variance. Differences between corticosteroid-treated and non-corticosteroid-treated groups with regard to mortality and perching ability were analyzed with use of the Fisher exact test. An alpha value of 0.05 was considered significant.
Source of Funding
One of the funding sources for this study, the Howard Holderness Distinguished Medical Scholars Program, provided part of the research salary for one of the authors (R.W.D.). The remaining two sources of funding for this project, the Aileen Stock Orthopaedic Research Fund and the Alpha Omega Alpha Carolyn L. Kuckein Student Research Fellowship, funded the purchase of experimental supplies and research animals. These sources were also used to cover animal-care fees. All sources of funding played no role in the investigation.
The differences between the experimental and control-toe active flexion measurements (the sum of active flexion measurements at the proximal and middle interphalangeal joints) for each experimental group at each time point are displayed in Table II. As illustrated in Table II, there were no significant differences in any of the six treatment groups between the experimental and control toes in terms of flexion prior to inoculation. At twenty-four hours following inoculation, immediately prior to treatment, the experimental toes all had significantly less active flexion in comparison with the matched contralateral, control toes.
At seven days following treatment, the difference in active flexion between the experimental toe and the matched contralateral control toe was <10° for all three groups receiving corticosteroids but significant differences were still found in all groups except the experimental group receiving local antibiotics, no surgical drainage, and corticosteroids (LABX+, SURG—, STER+), which had a mean difference of only 4° ± 2° (p = 0.05).
At fourteen days following treatment, the difference between active flexion in the experimental toe and in the matched contralateral, control toe had increased to >10° for all groups except for that same group (the group receiving local antibiotics, no surgical drainage, and corticosteroids [LABX+, SURG—, STER+]), which had a mean difference of only 2° ± 4° (p = 0.57).
By twenty-eight days following treatment, the three corticosteroid-treated groups as well as the group treated with local antibiotics, no surgical drainage, and no corticosteroids (LABX+, SURG—, STER—) had <10° of difference in active flexion between the experimental and control toes. Significant differences were still detected in the group treated with local antibiotics, surgical drainage, and corticosteroids (LABX+, SURG+, STER+) and the group treated with local antibiotics, no surgical drainage, and no corticosteroids (LABX+, SURG—, STER—).
The differences in active flexion between the experimental and control toes were compared between groups as well. All groups were compared with the current standard of care (systemic antibiotics, surgical drainage with catheter irrigation, and no corticosteroids [SABX+, SURG+, STER—]). Figure 2 is a graphic representation of the percentage decrease in active flexion between the experimental toes and the matched control toes and the significance of these differences for each of the groups as compared with the standard of care.
Compared with the standard-of-care treatment (systemic antibiotics, surgical drainage with catheter irrigation, and no corticosteroids [SABX+, SURG+, STER—]), several groups regained significantly more flexion at various post-treatment time points (Table III). The group with the addition of corticosteroids to the current standard of care (SABX+, SURG+, STER+) did not perform significantly better than the standard-of-care group (SABX+, SURG+, STER—), although it nearly reached significance at Post-Treatment Day 7 (9° ± 5°, p = 0.05) and at Post-Treatment Day 28 (11° ± 6°, p = 0.07). The group treated with local antibiotics, no surgical drainage, and corticosteroids (LABX+, SURG—, STER+) regained significantly more flexion than the standard-of-care group on Post-Treatment Day 7 (12° ± 4°, p < 0.05), Day 14 (14° ± 5°, p < 0.05), and Day 28 (12° ± 6°, p < 0.05). The group treated with local antibiotics, no surgical drainage, and no corticosteroids (LABX+, SURG—, STER—) also performed better than the standard-of-care group by Post-Treatment Day 28 (11° ± 5°, p < 0.05), as did the group treated with local antibiotics, surgical drainage, and corticosteroids (LABX+, SURG+, STER+) (12° ± 5°, p < 0.05).
Analysis of variance was used to compare the effects of groups of treatment on the recovery of active flexion. Comparing the differences between all experimental and control toes treated with corticosteroids and all those not treated with corticosteroids revealed that the corticosteroid-treated groups regained significantly more active flexion than did the non-corticosteroid-treated groups on Post-Treatment Day 7 (9° ± 2°, p < 0.05), Day 14 (7° ± 3°, p < 0.05), and Day 28 (6° ± 3°, p < 0.05) (Fig. 3). Also, all surgically treated groups regained function at a slower rate than all non-surgically treated groups, with the non-surgically treated groups regaining significantly more flexion at Post-Treatment Day 7 (by 5° ± 2°, p < 0.05), and Day 14 (by 8° ± 3°, p < 0.05). By Post-Treatment Day 28, both surgically and non-surgically treated groups had regained similar active flexion (p = 0.12). At no time point did the active flexion regained by the systemically administered antibiotic and locally administered antibiotic groups differ significantly.
The perching measurements revealed that eighty-four of eighty-seven chickens were able to perch on the smallest perch prior to inoculation. Twenty-four hours following inoculation, perching ability was greatly impaired across all experimental groups, with only 26% (twenty-three) of eighty-seven chickens being able to perch on the smallest perch. Figure 4 shows the percentage of chickens in pooled groups treated with corticosteroids (STER+) and treated without corticosteroids (STER—) that were able to perch on the smallest perch (through maximum flexion) on Post-Treatment Days 7, 14, and 28. At all time points, but most notably on Post-Treatment Day 7 (85.4% [thirty-five of forty-one chickens] in the corticosteroid group, compared with 41.5% [seventeen of forty-one chickens] in the non-corticosteroid group), a higher percentage of chickens treated with corticosteroids, regardless of other treatment methods, were able to achieve maximum active flexion and to successfully perch on the smallest-diameter perch. Significantly more of the chickens that were treated with corticosteroids were able to perch on the smallest perch on Post-Treatment Day 7 in comparison with chickens that were treated without corticosteroids (p < 0.05). The differences in perching ability between corticosteroid and non-corticosteroid groups were not significantly different on Post-Treatment Day 14 (p = 0.26) or Day 28 (p = 0.46).
Not all chickens survived the entire experiment. All groups had at least one chicken that either died or was killed during the course of the experiment. The most common reasons for chickens to be killed were lameness or weight loss, which were judged by the veterinary staff to be secondary to sepsis in all cases. The group treated with systemic antibiotics, surgical drainage, and no corticosteroids (SABX+, SURG+, STER—); the group treated with systemic antibiotics, surgical drainage, and corticosteroids (SABX+, SURG+, STER+); and the group treated with local antibiotics, surgical drainage, and corticosteroids (LABX+, SURG+, STER+) all lost one chicken during the course of the experiment (on Post-Inoculation Days 8 [killed], 9 [killed], and 2 [died], respectively). The group treated with local antibiotics, surgical drainage, and no corticosteroids (LABX+, SURG+, STER—) lost three chickens during the course of the experiment (on Post-Inoculation Days 10 [killed], 17 [died], and 17 [died]), whereas the group treated with local antibiotics, no surgical drainage, and no corticosteroids (LABX+, SURG—, STER—) and the group treated with local antibiotics, no surgical drainage, and corticosteroids (LABX+, SURG—, STER+) lost two chickens during the course of the experiment (on Post-Inoculation Days 21 [killed] and 18 [killed] and on Days 15 [died] and 13 [killed], respectively). With use of the Fisher exact test, no significant difference was found between the number of chickens that died or were killed between groups of chickens treated with or without corticosteroids (p = 0.52).
The progression of pyogenic flexor tenosynovitis can be rapid and can result in serious long-term stiffness of the infected digit if not promptly diagnosed and treated. The current standard of care for the treatment of this infection is surgical irrigation and debridement and appropriate systemic antibiotic coverage. As corticosteroids have been shown to effectively decrease morbidity in cases of other closed-space infections such as septic arthritis, infectious nephritis, and bacterial meningitis, we hypothesized that a serious closed-space infection of the hand, pyogenic flexor tenosynovitis, might also respond favorably to the addition of local corticosteroids to the treatment regimen14-19.
Our data support our hypothesis that the administration of local corticosteroids as an adjunct to antibiotic treatment of pyogenic flexor tenosynovitis in a broiler chicken allows significantly greater recovery of digital range of motion (compared with controls) with or without surgical drainage. This recovery of range of motion is likely secondary to the effect of corticosteroids, which decreases the inflammatory cell recruitment that has been shown to be responsible for a great amount of the damage in septic arthritis20 and may be involved in the production of adhesions in pyogenic flexor tenosynovitis.
Although inherent limitations exist in any animal study, our model did have a number of strengths. Chickens have long been an animal model of choice for various flexor tendon pathologies because of similarities in structure between the flexor tendons of the human hand and those of the chicken. Both species possess similarly functioning flexor digitorum superficialis and flexor digitorum profundus tendons25-29. Also, broiler chickens have separate tendon sheaths for each toe, similar to the tendon-sheath structure of the human hand26,27. The large size of chicken flexor tendons also make them more amenable to surgical interventions than the flexor tendons of smaller mammalian species.
Similar to humans, chickens with pyogenic flexor tenosynovitis exhibit most of the Kanavel signs of pyogenic flexor tenosynovitis, including holding of the digit in a partially flexed posture, uniform and quantifiable swelling along the infected digit, and apparent tenderness on percussion of the digit30. The fourth Kanavel sign, pain on extension of the digit, was not evident in all chickens, although some chickens did walk with a marked limp following infection, in an apparent attempt to avoid excessive digital extension and pressure on the digit30.
Although clinical trials are necessary to warrant any change in clinical practice in the treatment of pyogenic flexor tenosynovitis, our study raises questions regarding the effectiveness of the current standard of care. Nearly all groups that were treated with corticosteroids regained significantly more active flexion than the standard-of-care group by Post-Treatment Day 28. Additionally, the group treated with local antibiotics alone regained significantly more active flexion than did the standard-of-care group.
Although all groups that received corticosteroids experienced significantly increased digital range of motion in comparison with groups that did not receive corticosteroids, the group treated with local antibiotics and corticosteroids without surgical drainage (LABX+, SURG—, STER+) fared exceptionally well at all post-treatment time points, with no significant difference noted between the amount of total flexion of the treated digit and that in the contralateral, uninfected, control digit. That same group (LABX+, SURG—, STER+) was also the only group that regained significantly more digital flexion than did the standard-of-care treatment group (SABX+, SURG+, STER—) at all post-treatment time points. This result suggests that surgical drainage may not be necessary for the treatment of this infection if a timely local injection of antibiotics into the tendon sheath is given in place of surgery. This finding is corroborated by the fact that the only non-steroid-treated group to regain significantly more active digital flexion than the standard-of-care group was the group treated with local antibiotics, no surgical drainage, and no corticosteroids (LABX+, SURG—, STER—), which regained significantly more flexion than the standard-of-care group by Post-Treatment Day 28. It is likely that even in the absence of infection, the inflammatory response elicited by the healing surgical wound leads to increased digital stiffness.
Local antibiotic delivery for the treatment of closed-space infections is not unique to this experiment as it has been suggested in the veterinary literature for the treatment of pyogenic flexor tenosynovitis and septic arthritis in horses21. Intra-articular antibiotics also have been used to treat septic arthritis in humans22. Local administration of antibiotics into surgical wounds results in significantly decreased wound bacterial counts as compared with those for wounds treated only with systemic antibiotics, possibly because of the ability of local antibiotics to be delivered at higher concentrations that would result in toxicity if given systemically23. To our knowledge, very little has been published on the host-tissue response to locally delivered antibiotics. The few animal studies on this topic have shown mixed results with regard to whether or not intra-articular antibiotic administration causes an increased host inflammatory response31,32. However, one study demonstrated that these inflammatory changes were transient and resolved within seven days after local antibiotic injection33. We postulate that a one-time dose of local antibiotics, injected directly into the tendon sheath in cases of pyogenic flexor tenosynovitis, may effectively sterilize the tendon sheath and prevent further bacterially mediated damage to the flexor tendon and that the needle puncture wound may to some extent drain out purulent material that is under pressure. However, a sterile tendon sheath filled with dead bacteria would be expected to elicit a strong host inflammatory response. Locally delivered corticosteroids could possibly dampen this inflammatory response and prevent host-mediated damage even after resolution of the infection. It must be noted that, in the present study, we were certain that the infecting organism was sensitive to the antibiotic used. It is encouraging to note that six chickens in the non-corticosteroid groups and four in the corticosteroid groups died or were killed because of sepsis and that this number of fatalities was not significantly different between groups, suggesting that corticosteroids did not result in more severe infections. However, it may be quite dangerous to administer corticosteroids if the bacterial organism might be resistant to, or incompletely or inappropriately treated with, the chosen antibiotic regimen.
The present study had some limitations. As it was an animal study, we cannot state that the same treatment regimen would effectively decrease loss of range of motion resulting from pyogenic flexor tenosynovitis in humans. Another possible weakness of the study could be the method of irrigation used during the surgical drainage procedure. Although some controversy over the preferred method of irrigation of pyogenic flexor tenosynovitis still exists in the literature, indwelling irrigation catheters providing serial wound irrigations have been found to be effective for the irrigation of pyogenic flexor tenosynovitis1. Given the bipedal nature of chickens and their inability to cooperate with interventions, leaving an indwelling catheter in the foot of the surgically drained chickens was not feasible as these catheters likely would have been removed by the chickens or would have become contaminated with feces. Similarly, intramuscular antibiotic administration was chosen over intravenous antibiotic administration because maintenance of intravenous access in chickens was not feasible. Although intravenous antibiotics are judged to be the standard of care for the treatment of pyogenic flexor tenosynovitis in humans, data on broiler chickens have indicated that intravenous and intramuscular antibiotics have nearly identical pharmacokinetic properties33. Thus, there is a very low likelihood that choosing intramuscular over intravenous antibiotic administration would skew our data.
Another weakness of the study was the method of measurement of active flexion with a goniometer, which could be subject to measurement error and inaccuracy. However, given the blinded manner of measurement, it is unlikely that possible measurement error affected one experimental group more than another. Yet another weakness of the study could be that the perching ability of the chickens could have been inhibited by pain, especially at twenty-four hours post-inoculation and at early post-treatment time points, possibly affecting the reliability of these data. However, although these are categorical data, we do believe that they add a valuable, more clinically relevant outcome to the study.
An additional weakness could lie in the time frame of the study. All chickens were killed on Post-Treatment Day 28 after appropriate measurements of flexion were taken. Given the known complication of tendon rupture in the setting of chronic corticosteroid exposure, it is possible that the tendons lying within the tendon sheaths treated with corticosteroid injections for pyogenic flexor tenosynovitis in humans could be at higher risk for rupture. This would be an interesting and necessary question to address with future animal studies prior to clinical trials to investigate corticosteroids as an adjunct treatment for pyogenic flexor tenosynovitis.
In summary, in our chicken model of pyogenic flexor tenosynovitis, the addition of locally administered corticosteroids to the treatment regimen significantly improved active digital flexion compared with chickens that did not receive local corticosteroids at all post-treatment time points. Specifically, locally administered corticosteroids in combination with locally administered antibiotics without surgical drainage resulted in the earliest and most dramatic recovery of active flexion following pyogenic flexor tenosynovitis. However, all corticosteroid-treated groups regained significantly more active digital flexion in the infected digit than did the non-corticosteroid-treated groups. Nearly all of the corticosteroid-treated groups regained significantly more active digit flexion than did the standard-of-care treatment group that received systemic antibiotics and surgical drainage. Locally injected antibiotics alone also led to significantly more active digital flexion than did the standard-of-care treatment.
Our data suggest that pyogenic flexor tenosynovitis behaves in a similar manner to other closed-space infections such as septic arthritis. Clinical trials are warranted to determine whether or not local corticosteroids should be considered as an adjunct treatment option to appropriately tailored antibiotic therapy and possible surgical drainage and irrigation for pyogenic flexor tenosynovitis in humans. Also, locally delivered antibiotics could represent an adjunctive treatment modality for pyogenic flexor tenosynovitis, perhaps decreasing the need for surgical drainage.
Note: The present study was supported by an Alpha Omega Alpha Carolyn L. Kuckein Student Research Fellowship, the Aileen Stock Orthopaedic Research Fund, and the Howard Holderness Distinguished Medical Scholars Program.
Boles
SD;
Schmidt
CC. Pyogenic flexor tenosynovitis. Hand Clin.
1998;14:567-78.
[PubMed]
Abrams
RA;
Botte
MJ. Hand infections: treatment recommendations for specific types. J Am Acad Orthop Surg.
1996;4:219-30.
[PubMed]
Pang
HN;
Teoh
LC;
Yam
AK;
Lee
JY;
Puhaindran
ME;
Tan
AB. Factors affecting the prognosis of pyogenic flexor tenosynovitis. J Bone Joint Surg Am.
2007;89:1742-8.
[PubMed][CrossRef]
Gutowski
KA;
Ochoa
O;
Adams
WP
Jr. Closed-catheter irrigation is as effective as open drainage for treatment of pyogenic flexor tenosynovitis. Ann Plast Surg.
2002;49:350-4.
[PubMed]
Spiegel
JD;
Szabo
RM. A protocol for the treatment of severe infections of the hand. J Hand Surg Am.
1988;13:254-9.
[PubMed]
Clark
DC. Common acute hand infections. Am Fam Physician.
2003;68:2167-76.
[PubMed]
Juliano
PJ;
Eglseder
WA. Limited open-tendon-sheath irrigation in the treatment of pyogenic flexor tenosynovitis. Orthop Rev.
1991;20:1065-9.
[PubMed]
Neviaser
RJ. Closed tendon sheath irrigation for pyogenic flexor tenosynovitis. J Hand Surg Am.
1978;3:462-6.
[PubMed]
Neviaser
RJ. Tenosynovitis. Hand Clin.
1989;5:525-31.
[PubMed]
Lille
S;
Hayakawa
T;
Neumeister
MW;
Brown
RE;
Zook
EG;
Murray
K. Continuous postoperative catheter irrigation is not necessary for the treatment of suppurative flexor tenosynovitis. J Hand Surg Br.
2000;25:304-7.
[PubMed]
Nemoto
K;
Yanagida
M;
Nemoto
T. Closed continuous irrigation as a treatment for infection in the hand. J Hand Surg Br.
1993;18:783-9.
[PubMed]
Bauman
JT;
Millon
SJ;
Tanner
SL. The outpatient treatment of pyogenic flexor tenosynovitis. J Surg Orthop Adv.
2005;14:92-5.
[PubMed]
Hausman
MR;
Lisser
SP. Hand infections. Orthop Clin North Am.
1992;23:171-85.
[PubMed]
Sakiniene
E;
Bremell
T;
Tarkowski
A. Addition of corticosteroids to antibiotic treatment ameliorates the course of experimental Staphylococcus aureus arthritis. Arthritis Rheum.
1996;39:1596-605.
[PubMed]
Wysenbeek
AJ;
Volchek
J;
Amit
M;
Robinson
D;
Boldur
I;
Nevo
Z. Treatment of staphylococcal septic arthritis in rabbits by systemic antibiotics and intra-articular corticosteroids. Ann Rheum Dis.
1998;57:687-90.
[PubMed]
Odio
CM;
Ramirez
T;
Arias
G;
Abdelnour
A;
Hidalgo
I;
Herrera
ML;
Bolaños
W;
Alpízar
J;
Alvarez
P. Double blind, randomized, placebo-controlled study of dexamethasone therapy for hematogenous septic arthritis in children. Pediatr Infect Dis J.
2003;22:883-8.
[PubMed]
Ho
G
Jr;
Su
EY. Antibiotic therapy of septic bursitis. Its implication in the treatment of septic arthritis. Arthritis Rheum.
1981;24:905-11.
[PubMed]
Verba
V;
Sakiniene
E;
Tarkowski
A. Beneficial effect of glucocorticoids on the course of haematogenously acquired Staphylococcus aureus nephritis. Scand J Immunol.
1997;45:282-6.
[PubMed]
van de Beek
D;
de Gans
J;
McIntyre
P;
Prasad
K. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev.
2007;1:.
Tarkowski
A;
Wagner
H. Arthritis and sepsis caused by Staphylococcus aureus: can the tissue injury be reduced by modulating the host's immune system?Mol Med Today.
1998;4:15-8.
[PubMed]
Schneider
RK;
Bramlage
LR;
Mecklenburg
LM;
Moore
RM;
Gabel
AA. Open drainage, intra-articular and systemic antibiotics in the treatment of septic arthritis/tenosynovitis in horses. Equine Vet J.
1992;24:443-9.
[PubMed]
Jerosch
J;
Hoffstetter
I;
Schröder
M;
Castro
WH. Septic arthritis: arthroscopic management with local antibiotic treatment. Acta Orthop Belg.
1995;61:126-34.
[PubMed]
Yarboro
SR;
Baum
EJ;
Dahners
LE. Locally administered antibiotics for prophylaxis against surgical wound infection. An in vivo study. J Bone Joint Surg Am.
2007;89:929-33.
[PubMed]
Houshian
S;
Seyedipour
S;
Wedderkopp
N. Epidemiology of bacterial hand infections. Int J Infect Dis.
2006;10:315-9.
[PubMed]
Jaibaji
M;
Brody
GS;
Rodgers
K;
Espinoza
T;
Roda
N;
Maldonado
S;
Pines
E;
diZerega
G. A new model for experimental tendon adhesions in the chicken. Ann Plast Surg.
2000;44:205-10.
[PubMed]
Lindsay
WK;
Thomson
HG. Digital flexor tendons: an experimental study. Part I. The significance of each component of the flexor mechanism in tendon healing. Br J Plast Surg.
1960;12:289-316.
[PubMed]
Farkas
LG;
Thomson
HG;
Martin
R. Some practical notes on the anatomy of the chicken toe for surgeon investigators. Plast Reconstr Surg.
1974;54:452-8.
[PubMed]
Strauch
B;
de Moura
W;
Ferder
M;
Hall
C;
Sagi
A;
Greenstein
B. The fate of tendon healing after restoration of the integrity of the tendon sheath with autogenous vein grafts. J Hand Surg Am.
1985;10(
6 Pt 1):790-5.
[PubMed]
Tang
JB;
Wang
YH;
Gu
YT;
Chen
F. Effect of pulley integrity on excursions and work of flexion in healing flexor tendons. J Hand Surg Am.
2001;26:347-53.
[PubMed]
Kanavel
AB. The symptoms, signs and diagnosis of acute suppurative tenosynovitis. In: Infections of the hand
. 6th ed. Philadelphia: Lea & Febiger; 1933. p 364-7.
Haerdi-Landerer
MC;
Suter
MM;
Steiner
A. Intra-articular administration of doxycycline in calves. Am J Vet Res.
2007:68:1324-31.
[PubMed]
Stover
SM;
Pool
RR. Effect of intra-articular gentamicin sulfate on normal equine synovial membrane. Am J Vet Res.
1985;46:2485-91.
[PubMed]
Abu-Basha
EA;
Idkaidek
NM;
Al-Shunnaq
AF. Comparative pharmacokinetics of gentamicin after intravenous, intramuscular, subcutaneous and oral administration in broiler chickens. Vet Res Commun.
2007;31:765-73.
[PubMed]