Septic Tenosynovitis of the Hand

(1)

Article

Reference

Septic Tenosynovitis of the Hand

MÜLLER, Camillo T., et al.

Abstract

Treatment of septic hand tenosynovitis is complex, and often requires multiple débridements and prolonged antibiotic therapy. The authors undertook this study to identify factors that might be associated with the need for subsequent débridement (after the initial one) because of persistence or secondary worsening of infection. METHODS: In this retrospective single-center study, the authors included all adult patients who presented to their emergency department from 2007 to 2010 with septic tenosynovitis of the hand.

MÜLLER, Camillo T., et al. Septic Tenosynovitis of the Hand. Plastic and Reconstructive Surgery, 2015, vol. 136, no. 3, p. 338e-343e

DOI : 10.1097/PRS.0000000000001510 PMID : 26313838

Available at:

http://archive-ouverte.unige.ch/unige:90592

Disclaimer: layout of this document may differ from the published version.

(2)

S

eptic flexor tenosynovitis (phlegmona) of the hand is a frequent reason for emergency hand surgery.^1,2 It can affect people of all ages^3,4 and has a variety of causes, such as cuts,

punctures by toothpick⁵ or fishbone,^6,7 or other injuries. Hallmarks of its management include urgent surgical débridement and prolonged antibiotic therapy.⁸ In a substantial number of cases, a subsequent surgical intervention (second-look surgery)^3,9 is required to avoid such complications as persistent or recurrent infection,¹⁰ stiff digits,¹¹ or amputations.^3,11,12 Optimal management of this condition requires the identification of patients at risk for, and potentially preventing, these complications.³ There is little published literature regarding clinical variables that may predispose to the need for a second-look operative procedure.

We therefore undertook a single-center study to investigate which factors (including duration of antibiotic therapy) in patients with hand phlegmona and septic tenosynovitis might be associated with the need for a second-look operation.

Disclosure: The authors have no conflicts of interest or funding in connection with this work.

Camillo T. Müller, M.D.

Ilker Uçkay, M.D.

Paolo Erba, M.D.

Benjamin A. Lipsky, M.D.

Pierre Hoffmeyer, M.D.

Jean-Yves Beaulieu, M.D.

Geneva and Lausanne, Switzerland;

and Oxford, United Kingdom

Background: Treatment of septic hand tenosynovitis is complex, and often requires multiple débridements and prolonged antibiotic therapy. The authors undertook this study to identify factors that might be associated with the need for subsequent débridement (after the initial one) because of persistence or secondary worsening of infection.

Methods: In this retrospective single-center study, the authors included all adult patients who presented to their emergency department from 2007 to 2010 with septic tenosynovitis of the hand.

Results: The authors identified 126 adult patients (55 men; median age, 45 years), nine of whom were immunosuppressed. All had community-acquired infection; 34 (27 percent) had a subcutaneous abscess and eight (6 percent) were febrile. All underwent at least one surgical débridement and had concomitant antibiotic therapy (median, 15 days; range, 7 to 82 days). At least one additional surgical intervention was required in 18 cases (median, 1.13 interventions; range, one to five interventions). All but four episodes (97 percent) were cured of infection on the first attempt after a median follow-up of 27 months. By multivariate analysis, only two factors were significantly associated with the outcome “subsequent surgical débridement”: abscess (OR, 4.6;

95 percent CI, 1.5 to 14.0) and longer duration of antibiotic therapy (OR, 1.2;

95 percent CI, 1.1 to 1.2).

Conclusion: In septic tenosynovitis of the hand, the only presenting factor that was statistically predictive of an increased risk of needing a second débride- ment was the presence of a subcutaneous abscess. (Plast. Reconstr. Surg. 136:

338e, 2015.)

CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.

From the Hand Surgery Unit, Department of Orthopaedic Sur- gery, Service of Infectious Diseases, Geneva University Hospi- tals and Faculty of Medicine, University of Geneva; and the Department of Infectious Diseases, University of Oxford.

Received for publication October 14, 2014; accepted March 9, 2015.

The first two authors contributed equally to this publication.

This paper has been presented orally and as a poster at the 19th Congress of the Federation of European Societies for Surgery of the Hand, in Paris, France, June 18 through 21, 2014; the 12th Quadrennial Meeting of the European Society of Plastic Reconstructive and Aesthetic Surgery, in Edinburgh, United Kingdom, July 6 through 11, 2014; and 75th Annual Congress of the Swiss Society for Orthopaedics and Traumatology, in St. Gallen, Switzerland, June 24 through 26. 2015.

Septic Tenosynovitis of the Hand: Factors

Predicting Need for Subsequent Débridement

HAND/PERIPHERAL NERVE

(3)

Volume 136, Number 3 • Débridement in Hand Septic Tenosynovitis

339e PATIENTS AND METHODS

By review of the medical records in our retrospective cohort, we identified all adult patients hospitalized for infectious tenosynovitis and/or hand phlegmona in the Geneva University Hos- pital between January of 2007 and December of 2010. In this review, three of the authors (C.T.M., P.E., and I.U.) assessed 83 key variables regarding patient demographics, types of treatment, and clinical outcomes. Because of their atypical fea- tures, we excluded cases caused by gonococcal, fungal, nocardial, and mycobacterial infection;

recurrent episodes; pediatric cases; tenosynovitis with underlying septic arthritis or osteomyelitis;

or any inflammation related to rheumatic or crys- talline disease. The local hospital ethics commit- tee approved the study and waived the need for retroactive informed consent because this was a retrospective study of our own patients.

Septic tenosynovitis was defined as the presence of at least two of the following Kanavel signs¹³ on concomitant assessment by one resi- dent (in training) physician and one attending hand surgeon: (1) slight flexion of the affected digits; (2) fusiform swelling over the affected tendon; (3) tenderness over the affected tendon; and (4) pain on passive extension of the affected digit.

We required that these signs be coupled with clinical signs of hand infection, such as redness (or lymphangitis), warmth, fever, or loss of nor- mal function.

Surgical and Antibiotic Therapy in Our Center Surgery was performed under regional anes- thesia (brachial plexus block) with a tourniquet.

Antibiotics were withheld until after intraoperative microbiological samples were obtained. When the infection was localized to the tendon sheath (clas- sified as Michon stage I), a limited incision with drainage and irrigation was usually performed, as described by Neviaser.^14,15 In more advanced cases presenting with subcutaneous purulence (Michon stage II),¹⁶ a wider approach was pre- ferred. This consisted of performing a Bruner¹⁷ incision proximally and distally from the infection site and then excising the tendon sheath. In severe cases (Michon stage III),^16,18,19 the entire tendon sheath and all necrotic tendons were removed. All wounds were irrigated with antiseptic-free saline solution (0.9%), left open, and dressed with par- affin-impregnated gauze. There was no indwelling irrigation catheter used. Postoperatively, the hand was splinted continuously and elevated, and sub- mitted to daily chlorhexidine irrigation and dress- ing changes. At 48 to 72 hours postoperatively, a

decision was made for either secondary closure or further débridement. A second-look and delaying closure was usually selected in case of persistent or worsening purulence. Initial antibiotic therapy was administered intravenously, with either cefuroxime or amoxicillin-clavulanate. Antibiotic therapy was switched to oral agents according to the susceptibility testing and agreement between the treating surgeon and the dedicated infectious diseases specialist of the orthopedic service.²⁰ Statistical Analysis

The primary objective was to determine whether there are any clinical variables predictive of requiring a second-look débridement for persistent worsening infection. Because there were only a small number of infectious recurrences, we could not analyze this potential outcome. We performed group comparisons of our retrospective cohort using the Pearson chi-square, Fisher’s exact, or Wilcoxon rank sum test, as appropriate, and a logistic regression analysis with the outcome

“subsequent surgical débridement,” adjusted for case mix in a case-control design. In a step- wise fashion, we introduced variables, particu- larly including antibiotic-related variables, with a univariate value of p < 0.02 into the multivari- ate model. We checked all independent variables for confounding, colinearity, and interaction. We used STATA software version 9.0 (StataCorp, Col- lege Station, Texas), and for all comparisons we considered values of p < 0.05 (two-tailed) to be sta- tistically significant.

RESULTS

We found 126 cases of hand tenosynovitis in 126 adult patients (55 men; median age, 45 years;

range, 17 to 92 years). Among these patients, nine (7 percent) had a potentially immunocom- promised status from diabetes mellitus (n = 5), dialysis (n = 1), autoimmune disease requiring corticosteroid therapy (n = 1), advanced cirrho- sis (n = 1), and untreated human immunodefi- ciency virus disease (n = 1). All infections were community acquired, including 20 caused by bite wounds. Osteosynthesis materiel was pres- ent in 15 cases: Kirschner wires in 10 cases and plates in five cases. Infections involved the right hand in 65 cases (41 percent); the middle finger was affected in 49 cases (39 percent), the index finger was affected in 41cases (33 percent), the thumb was affected in 25 cases (20 percent), the ring finger was affected in 21 cases (17 percent), and the little finger was affected in 15 cases

(4)

(12 percent). In 108 episodes (86 percent), only one digit was infected; in 11 episodes (9 percent), two digits were infected; in five episodes (4 percent), three digits were infected; and in two episodes (2 percent), all five digits of the same hand were infected. In 34 cases (27 percent), infection was associated with a subcutaneous abscess. The Kanaval signs¹³ included fusiform swelling [n = 90 (71 percent)], tenderness along the flexor sheath [n = 68 (53 percent)], pain on passive extension [n = 61 (48 percent)], and digit in flexed posture [n = 21 (17 percent)]. Among 124 patients in whom a Michon classification¹⁹ could be applied, 29 (23 percent) were Michon stage 0, lacking the copious amounts of synovial liquid needed for a Michon stage, 49 cases (38 percent) were Michon stage I, 37 cases (30 percent) were Michon stage II, and nine episodes (7 percent) were Michon stage III. Ten patients (8 percent) presented with lymphangitis, and eight (6 percent) were febrile (>38°C). Median C-reactive protein serum level was 20 mg/liter. The three most frequently iso- lated pathogens were Staphylococcus aureus in 35 cases (28 percent), Pasteurella multocida in nine cases (7 percent), and Streptococcus pyogenes in nine cases (7 percent) cases, whereas in 15 cases (12 percent), infection was polymicrobial.

Treatment and Outcomes

All patients underwent at least one surgical débridement and all received antibiotic therapy. The median delay between the causative trauma and surgery was 3 days. Eighteen patients (14 percent) underwent an additional subsequent débridement, each of which was at the initial operative site; no patient developed infection propagation to other digits. Three patients (2 percent) required two interventions, two (2 percent) had three interventions, two (2 percent) had four operations, and one (1 percent) underwent five débridements.

Overall, these patients were treated with 24 different regimens of antimicrobial therapy. Eigh- teen patients (14 percent) were already receiving antibiotic therapy at the time of hospital admission. For treating these infections, the median duration of antibiotic therapy by the intravenous route was 3 days (range, 0 to 55 days), and the median duration of total antibiotic therapy was 15 days (range, 7 to 82 days). The median length of hospital stay was 4 days (range, 1 to 36 days), and was it was significantly prolonged in those with more than one surgical intervention (4 days versus 15 days) (Table 1). After their discharge from the

hospital, we followed our patients with a median of six consultations and a median active follow-up of 27 months (range, 9 to 63 months). Only four patients had a true infection recurrence, defined as infection of the same site, with the same patho- gen, without new trauma. These occurred after a median duration of 6 months after the cessation of treatment for the first episode. Five patients (4 percent) required reconstructive surgery to regain satisfactory digit function, and one necrotic digit was amputated for mechanical reasons.

Multivariate Results

In view of differences in the crude group comparisons (Table 1), we performed a multivariate analysis to adjust for case mix (Table 2). Only two variables were found to have a statistically significant association with the outcome subsequent surgical débridement: abscess (OR, 4.6; 95 percent CI, 1.50 to 14.0) and total duration of antibiotic administration (OR, 1.17; 95 percent CI, 1.05 to 1.19). The result of the goodness-of-fit test was not statistically significant (p = 0.54) and the receiver operating characteristic curve value was 0.84, sug- gesting acceptable accuracy of our final model.

DISCUSSION

Our retrospective study found that in patients with septic hand tenosynovitis, the presence of an abscess at initial evaluation was the only predictive factor associated with a need for subsequent débridement (i.e., a second-look operation). In contrast, various demographic factors (e.g., age, sex), clinical findings (e.g., presence of immune suppression, phlegmona classification schemes, number of fingers involved), laboratory results (e.g., C-reactive protein levels, pathogens isolated), and the origin of infection were all unrelated to patients needing another surgical procedure. Likewise, receiving treatment with an antibiotic before undergoing débridement was not associated with the need for a second-look procedure.

We can postulate two possible explanations for these findings. First, as the septic tenosynovitis progresses, the pus distends the flexor sheath,²¹ which may then rupture and allow the spread of pus into the subcutaneous planes.²¹ The rapid accumulation of pus also increases tissue compartment pressure, further compromising blood sup- ply and potentially causing blistering, necrosis, and gangrene.²¹ These complications in turn may also decrease the ability of antibiotics to penetrate to the site of infection. Consequently, subcutaneous

(5)

341e purulence requires more extensive débridement,

with an increased potential for delayed resolution and mobilization, adhesion formation, or ischemia of the digit.²¹ The combination of extensive débridement, delayed resolution, and tendon isch- emic damage, as in Michon stage III,¹⁹ may result in poor function of any surviving digits. Second, we observed that antibiotic therapy was significantly prolonged in patients who underwent multiple

débridements. This was probably attributable to confounding by indication, because the total duration of antibiotic therapy was unrelated to the likelihood of recurrence of infection. Despite the duration of antibiotic therapy ranging from 7 to 82 days for our patients, we noted only four infection recurrences (3 percent). Thus, we question the widespread practice of automatically prolonging antibiotic therapy in cases requiring reoperation.

Table 1. Comparison of the Patient Groups with and without Subsequent Débridement in Pyogenic Hand Tenosynovitis

Single Débridements (%) Comparison (p*) ≥2 Débridements (%)

No. 108 18

Patient characteristics

Male sex 55 (51) 13 (72)

Median age, yr 54

0.03 43

Immunosuppression† 6 (6) 3 (17)

Bites 17 (16) 3 (17)

Michon stage

0 26 (24) 3 (17)

I 43 (40) 6 (33)

II 31 (29) 6 (33)

III 6 (6) 3 (17)

Presence of foreign body 13 (12) 2 (11)

Antibiotic therapy before surgery 14 (13) 4 (22)

Causative pathogens

Staphylococcus aureus 29 (27) 6 (33)

Pasteurella multocida 8 (7) 1 (6)

Streptococcus pyogenes 7 (6) 2 (11)

Polymicrobial infection 9 (8) 0.00 6 (33)

Right hand 54 (50) 11 (61)

Dominant hand 43 (40) 7 (39)

Fingers involved

Thumb 20 (19) 5 (28)

Index 36 (33) 7 (39)

Middle 41 (38) 8 (44)

Ring 16 (15) 5 (28)

Little 12 (11) 3 (17)

No. of fingers involved

One 96 (89) 12 (67)

Two 7 (6) 4 (22)

Three 4 (4) 1 (6)

Five 1 (1) 1 (6)

Kanavel signs

Flexed posture 20 (19) 1 (6)

Fusiform swelling 76 (70) 14 (78)

Pain on passive extension 55 (51) 6 (33)

Tenderness along flexor sheath 59 (55) 9 (50)

Inflammatory signs

Lymphangitis 9 (8) 1 (6)

Abscess 23 (21) 0.00 11 (61)

Fever (>38°C) 6 (6) 2 (11)

Median CRP value on admission, mg/liter 14.5 0.02 32

Treatment

Operating surgeon’s experience

Chief surgeon 21 (19) 4 (22)

First attending surgeon 17 (16) 5 (28)

Second attending surgeon 20 (19) 3 (17)

Median duration of antibiotic therapy, days 15 0.00 30

Median duration IV antibiotic therapy, days 3 0.00 6.5

Outcomes

Median length of hospital stay, days 4 0.00 14.5

Median no. of consultations 6 0.00 10

CRP, C-reactive protein; IV, intravenous.

*Only significant p values (p ≤ 0.05, two-tailed) are displayed.

†Diabetes mellitus, steroids, dialysis, Child class C cirrhosis, and human immunodeficiency virus disease.

(6)

Previous studies have reported several factors that purportedly predicted the function of digits²² and the probability of digit amputation in patients treated for septic tenosynovitis.¹⁸ In 1974, Michon described a staging based on the presence or absence of three intraoperative findings:

synovial purulence, subcutaneous purulence, and necrosis of the flexor tendon.¹⁹ This classification has been used to help surgeons in selecting the operative technique and predicting the functional outcome.¹⁶ Michon suggested treating stage I cases with repeated irrigation with saline solution, stage II by incision of pulleys and tendons with explora- tion of interphalangeal and metacarpophalangeal joints, and stage III with excision of all necrotic tissue.¹⁹ The presence of an abscess was not part of this classification scheme. Sokolow et al. confirmed a correlation between the Michon classification and the final functional state of the involved digit.¹¹ More recently, in a study of 75 patients, Pang et al. assessed several variables, such as patient age or comorbidities, presence of an abscess or local ischemia, and the number of bacteria species growing on culture media. Their analysis revealed that age older than 43 years and having diabetes mellitus, an abscess or digital ischemia, or a polymicrobial infection are factors associated with the need for

digit amputation in septic tenosynovitis.³ However, they did not analyze the important outcome of the number of surgical débridements performed.

Unlike other groups,^3,14,23 we could not determine an association of diabetes with subsequent surgery or amputation, mostly because only one patient underwent amputation in one digit. Grin- nell²⁴ noted that the prevalence of tendon slough was higher in cases of polymicrobial infection.

Other reports have described cases of infection with mixed organisms, especially in immunocom- promised patients such as intravenous drug abus- ers, patients with diabetes mellitus, and patients treated with corticosteroids.^3,14,23 We found no evidence for an association between polymicrobial infection and the need for subsequent injury (p ≥ 0.05); however, with a larger cohort, this association might have achieved statistical significance.

CONCLUSIONS

In our large series of cases of septic flexor tenosynovitis, we found that only the presence of an abscess was statistically associated with the need for a subsequent surgical débridement.

The limitations of our study include its retrospective design, the fact that the infections were Table 2. Associations with Multiple Débridement of Tenosynovitis

OR (95% CI)

Univariate Analysis Multivariate Analysis

Age* 1.03 (1.00–1.05) 1.01 (0.98–1.04)

Male sex 2.51 (0.84–7.51)

Immunosuppression 3.40 (0.77–15.06) 0.35 (0.04–3.14)

Antibiotic therapy before surgery 1.91 (0.55–6.66)

Osteosynthesis material 0.91 (0.19–4.44)

Fever 2.13 (0.39–11.46)

CRP at admission* 1.01 (1.00–1.01) 1.00 (0.99–1.01)

Staphylococcus aureus 1.36 (0.47–3.96)

Pasteurella multocida 0.74 (0.09–6.26)

Streptococcus pyogenes 1.80 (0.34–9.46)

Polymicrobial infection 5.50 (1.67–18.15)† 3.76 (0.99–14.21)

Bites 1.07 (0.28–4.10)

Thumb 1.69 (0.54–5.29)

Index 1.27 (0.46–3.56)

Middle finger 1.48 (0.39–4.37)

Ring finger 2.21 (0.69–7.05)

Little finger 1.60 (0.40–6.34)

No. of fingers* 1.82 (1.02–3.22)† 1.39 (0.70–2.79)

Flexed posture 0.26 (0.03–2.06)

Fusiform swelling 1.47 (0.45–4.82)

Pain on passive extension 0.48 (0.17–1.38)

Tenderness along flexor sheath 0.83 (0.31–2.25)

Abscess 5.81 (2.02–16.66)† 4.60 (1.50–14.06)†

Michon degree 1.54 (0.87–2.73)

Delay from admission to surgery* 0.97 (0.89–1.08)

Duration of antibiotic therapy* 1.11 (1.06–1.16)† 1.17 (1.05–1.19)†

Duration of parenteral therapy* 1.40 (1.17–1.67)†

CRP, C-reactive protein.

*Continuous variable.

†Statistically significant (two-tailed value of p < 0.05).

(7)

343e community acquired,²⁵ and the fact that patients

were all immunocompetent adults from a single university center, thus limiting the generalizability of our findings to other settings or to the pediatric population. Furthermore, the treating surgeons decided based on clinical findings whether to reop- erate. This could lead to an overestimation bias in favor of a second intraoperative look. Patients with more impressive initial findings might therefore undergo an unnecessary second operation. We do not know what would occur if surgeons did not perform a second débridement for persistent pus and just left the wound open for local care and observation on the ward. Repeated surgery sub- stantially increases cost, often leads to the decision to continue (eventually unnecessary) antibiotic therapy, and prolongs the length of hospital stay.

Thus, the next step in developing an evidence- based treatment approach to this serious infection is to undertake a randomized trial to evaluate the necessity for subsequent débridement.

Camillo T. Müller, M.D.

Centre de la Main CHUV - University Hospital of Lausanne Avenue Pierre Decker 4, CH-1011 Lausanne, Switzerland camillo.muller@chuv.ch

ACkNOwLEDgMENT

The authors thank their colleagues of Hand Surgery and the Laboratory of Microbiology for assistance.

REFERENCES

1. Anwar MU, Tzafetta K, Southern SJ. Review of com- munity-referred hand infections. Surg Infect (Larchmt.) 2008;9:357–366.

2. Khundkar R. HOSTILE: A mnemonic for factors affect- ing hand infection outcome. J Plast Reconstr Aesthet Surg.

2011;64:838.

3. Pang HN, Teoh LC, Yam AK, Lee JY, Puhaindran ME, Tan AB. Factors affecting the prognosis of pyogenic flexor teno- synovitis. J Bone Joint Surg Am. 2007;89:1742–1748.

4. Luria S, Haze A. Pyogenic flexor tenosynovitis in children.

Pediatr Emerg Care 2011;27:740–741.

5. Chang MC, Huang YL, Liu Y, Lo WH. Infectious complica- tions associated with toothpick injuries of the hand. J Hand Surg Am. 2003;28:327–331.

6. Hudson DA, de Chalain TM. Hand infections secondary to fish bone injuries. Ann R Coll Surg Engl. 1994;76:99–101.

7. Young-Afat DA, Dayicioglu D, Oeltjen JC, Garrison AP.

Fishing-injury-related flexor tenosynovitis of the hand: A case report and review. Case Rep Orthop. 2013;2013:587176.

8. O’Malley M, Fowler J, Ilyas AM. Community-acquired meth- icillin-resistant Staphylococcus aureus infections of the hand:

Prevalence and timeliness of treatment. J Hand Surg Am.

2009;34:504–508.

9. Weinzweig N, Gonzalez M. Surgical infections of the hand and upper extremity: A county hospital experience. Ann Plast Surg. 2002;49:621–627.

10. Stern PJ, Staneck JL, McDonough JJ, Neale HW, Tyler G.

Established hand infections: A controlled, prospective study.

J Hand Surg Am. 1983;8:553–559.

11. Sokolow C, Dabos N, Lemerle JP, Vilain R. Bacterial flexor tenosynovitis in the hand: A series of 68 cases. Ann Chir Main 1987;6:181–188.

12. Evgeniou E, Iyer S. Pyogenic flexor tenosynovitis leading to an amputation. BMJ Case Rep. 2012;2012: pii: bcr2012006778.

13. Kanavel A. Infections of the Hand. Philadelphia: Lea & Febiger;

1912.

14. Mann RJ, Peacock JM. Hand infections in patients with dia- betes mellitus. J Trauma 1977;17:376–380.

15. Neviaser RJ. Closed tendon sheath irrigation for flexor teno- synovitis. J Hand Surg. 1978;3:462–466.

16. Kleinert HE. Should an incompletely severed tendon be sutured? Commentary. Plast Reconstr Surg. 1976;57:236.

17. Bruner JM. Surgical exposure of flexor tendons in the hand.

Ann R Coll Surg Engl. 1973;53:84–94.

18. McDonald LS, Bavaro MF, Hofmeister EP, Kroonen LT.

Hand infections. J Hand Surg Am. 2011;36:1403–1412.

19. Michon J. Phlegmon of the tendon sheaths (in French). Ann Chir. 1974;28:277–280.

20. Uçkay I, Vernaz-Hegi N, Harbarth S, et al. Activity and impact on antibiotic use and costs of a dedicated infectious diseases consultant on a septic orthopaedic unit. J Infect.

2009;58:205–212.

21. Schnall SB, Vu-Rose T, Holtom PD, Doyle B, Stevanovic M.

Tissue pressures in pyogenic flexor tenosynovitis of the fin- ger: Compartment syndrome and its management. J Bone Joint Surg Br. 1996;78:793–795.

22. Strickland JW, Glogovac SV. Digital function following flexor tendon repair in zone II: A comparison of immobilization and controlled passive motion techniques. J Hand Surg Am.

1980;5:537–543.

23. Kour AK, Looi KP, Phone MH, Pho RW. Hand infections in patients with diabetes. Clin Orthop Relat Res. 1996;331:238–244.

24. Grinnell RS. Acute suppurative tenosynovitis of the flexor tendon sheaths of the hand: A review of one hundred and twenty-five cases. Ann Surg. 1937;105:97–119.

25. Reber A, Moldovan A, Dunkel N, et al. Should the methicil- lin-resistant Staphylococcus aureus carriage status be used as a guide to treatment for skin and soft tissue infections? J Infect.

2012;64:513–519.