Concomitant with a rise in WALANT and a transition toward increased office-based procedures, the need for surgical antimicrobial prophylaxis has been called into question. Across several recent studies the infection rate following carpal tunnel release and similar procedures is consistently less than 1% regardless of the whether antibiotics are provided. In 454,987 Medicare patients undergoing carpal tunnel release, Werner et al reported a 0.32% infection rate.¹ A multistate, commercial insurance database study of 516,986 patients used propensity-score matching to find no difference in the risk for surgical site infection between patients who received antibiotic prophylaxis and those who did not.² These database studies have been corroborated by a large clinical study, which reported a 0.35% infection rate for clean, elective, hand surgeries performed at an outpatient surgical center. Again, no difference was found between patients who received antibiotics and those who did not.³ Some minor hand procedures may be safely performed in the emergency department or office with a similarly low infection rate.⁴ Finally, a randomized, double-blind study of 1,340 patients undergoing hand surgery found no decrease in infection rates with the administration of preoperative antibiotics. This led the authors to conclude that there is little evidence to support the routine use of preoperative antibiotics in hand surgery.⁵

Reported factors associated with the development of a surgical site infection include younger age, male sex, obesity, tobacco use, alcohol use, diabetes, and other medical comorbidities.^1,3 More important than the diagnosis of diabetes itself, however, may be the patient’s HbA1c level. Patients with a preoperative HbA1c greater than 8 are at a significantly increased risk for surgical site infection.⁶ Other relevant risks for infection include prolonged surgical time⁷ and delayed treatment for open hand fractures.⁸

A prospective study comparing surgical site preparation techniques found that prep-stick application led to more unprepared areas of skin compared with
immersed, sterile gauze sponges. The areas most at risk were distal to the PIP joint such as the hyponychium and finger pulp.⁹ A prospective randomized trial comparing preparation solutions found that the use of DuraPrep and Betadine was associated with significantly fewer positive cultures than ChloraPrep. However, the authors note that the majority of positive cultures identified were Bacillus species, a rare cause of postoperative infection.¹⁰

Whether to bury Kirschner wires and their effect on surgical site infection have been debated across multiple studies including two randomized controlled trials. A recently published retrospective series indicated that 7% of percutaneously placed K-wires were complicated by infection requiring intervention, but other groups have reported infection rates approaching 20%.¹¹ Among 695 patients treated with K-wires for fractures in the hand and wrist, exposed K-wires were more likely to be treated for pin-site infection (17.6%) than K-wires placed beneath the skin (8.7%).¹²

Hand involvement is an independent risk factor for hospital admission for patients presenting to the emergency department with cellulitis.¹³ Community-acquired MRSA (CA-MRSA) hand infections increased rapidly in prevalence over the past two decades, but may be decreasing as polymicrobial infections become more frequent.¹⁴ Anywhere from 30% to 70% of surgical hand infections are MRSA related. Prolonged hospitalization, chronic illness, IV drug abuse, and prior hand infection are known risk factors. The results of a systematic review recommend empiric coverage for CA-MRSA should be provided if local prevalence rates exceed 10% to 15%.¹⁵ Furthermore, a prospective randomized trial of patients presenting to a county emergency department diagnosed with MRSA hand infections demonstrated increased cost and mean hospital stay for patients treated with cefazolin compared with vancomycin. The trial was cut short when the incidence of MRSA infection in their study was found to be 72%.¹⁶ Alarmingly, multidrug-resistant strains of MRSA are being increasingly reported with a higher incidence of resistance to clindamycin and levofloxacin. Risk factors for such pathogens at an urban center included young age, intravenous drug use, and nosocomial infection.^14,17 This should be considered when selecting empiric antibiotic therapy.

For patients with fluctuance concerning for abscess, ultrasonography may be of value as a diagnostic test. Ultrasonography was shown to have a 78.4% positive predictive value of identifying an abscess and a negative predictive value of 90% to rule out an abscess in a series of 179 patients.¹⁸ Early antibiotic administration should be provided and has not been shown to greatly reduce bacterial culture growth from hand abscess so long as decompression is performed within 24 hours.¹⁹ After surgical decompression, débridement, and irrigation of the abscess, packing is often used, at least initially, to allow for continued drainage. No difference has been shown between different soaks and daily dressing changes in clearing the infection postoperatively.

Patients with immunosuppression are being increasingly encountered because of malignancies, transplants, or autoimmune disorders. In a matched cohort study, the most frequent immunosuppressive medication among patients with upper extremity infections was glucocorticoids. Infections in immunosuppressed patients were more likely to involve deeper structures such as joints, bone, tendons, and muscle.²⁰ Additionally, these patients are at a higher risk for atypical infections. They should be treated rapidly and aggressively as their potential for increased morbidity is high.

The clinical diagnosis of flexor tenosynovitis is classically made with the use of Kanavel signs. However, only slightly more than half of patients will exhibit all four signs and thus how best to apply these signs to patients suspected of having an infection is debated. Pang et al²¹ found that fusiform swelling was the most common sign (97% of patients) and that pain with passive extension of the digit was most specific (Figure 1). This is in contrast to a similar study by Kennedy et al which reported pain with passive extension to be the most sensitive sign and tenderness along the tendon sheath to be the most specific. Additionally, they identified three independent predictors of flexor tenosynovitis: tenderness along the sheath, pain with passive extension, and duration of symptoms less than 5 days. Patients with all three findings had an 87.9% likelihood of having the diagnosis compared with zero in patients with none of these signs or symptoms.²² It should also be noted that Kanavel signs are not uniformly present in children and adolescents.²³ Furthermore, inflammatory markers such as white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) are not sensitive enough to be used as a screening tool for ruling out flexor tenosynovitis in concerning patients. However, an elevation in any of these markers increases the likelihood of infection. The presence of a subcutaneous abscess on initial presentation is associated with poorer outcomes.²⁴

Patients who present with symptoms for less than 24 hours may undergo a trial of nonsurgical therapy with intravenous antibiotics. Ketonis et al²⁵ demonstrated in a cadaver study that S aureus readily forms thick biofilms on flexor tendons, suggesting the well-known role for surgical treatment. A variety of surgical approaches have been described. A recent systematic review reported excellent range of motion in 74% of patients treated with limited exposure of the sheath and closed catheter irrigation compared with 26% of patients treated with open surgical drainage.²⁶ The most common pathogens identified are S aureus and beta-hemolytic Streptococcus.²⁷ At a pediatric center, MRSA was the most commonly identified pathogen.²³ Infection with beta-hemolytic Streptococcus is associated with the need for more operations to eradicate infection and prolonged hospital stays. Rare pathogens in flexor tenosynovitis cultures recently reported include Shewanella algae,²⁸ Nocardia nova,²⁹ and Mycobacterium species.³⁰ Antibiotics are an important component of any surgical treatment. However, the ideal route of administration and duration has not been determined.²⁶ The use of catheter irrigation postoperatively has not been shown to improve outcomes in clinical studies.

Unfortunately, 14.2% patients will require at least one additional surgical intervention according to a recent retrospective, single-center study.²⁴ Worse outcomes are associated with older age, a delay in receiving antibiotics, and certain comorbidities.^21,31 In one series, 5 of 20 diabetic patients and all 3 patients on renal dialysis required an amputation because of complications from flexor tenosynovitis.²⁷

As few as 5% of patients presenting to the emergency department with an inflamed, atraumatic wrist joint will have a diagnosis of septic arthritis. In a review of 104 such patients at an academic medical center, a successful joint aspiration was able to be obtained in fewer than half.³² This emphasizes the importance of using the history and physical examination to determine the need for surgical intervention.

In a separate, retrospective comparison, patients with nonseptic arthritis were more likely to have chronic kidney disease, preexisting gout, or both compared with patients with confirmed septic arthritis. All infected patients had normal serum uric acid levels and two or more raised inflammatory markers with a mean CRP significantly higher compared with nonseptic group.³³ This finding is not consistent across studies and the possibility of a septic, gouty wrist does exist. However, diabetic and immunosuppressed patients may be more prone to develop septic arthritis.

Surgical treatments for septic arthritis include aspiration, arthroscopic débridement, or open arthrotomy with irrigation and débridement. A retrospective review of 36 patients reported increased repeat procedures required in patients initially treated with open versus arthroscopic irrigation and débridement.³⁴ After surgical drainage, a short course of intravenous antibiotics (less than 1 week) followed by oral antibiotics for 2 to 3 weeks successfully treated most cases of septic arthritis in the hand and wrist. S aureus is the most commonly isolated organism.

Approximately 600 to 1,200 cases of necrotizing fasciitis are reported each year in the United States.³⁵ When the infection is monomicrobial, it is most often caused by beta-hemolytic Streptococcus or Vibrio species. Polymicrobial infections are common and often involve aerobic and anaerobic organisms.

The initial diagnosis of necrotizing fasciitis is made on clinical examination in addition to available imaging studies and a constellation of laboratory results, such as the laboratory risk indicator for necrotizing fasciitis (LRINEC) score³⁶ (Table 1). Using a cutoff of 6 points, the positive predictive value of a patient having necrotizing fasciitis is 92%. The negative predictive value is 96%.³⁶ The utility of the LRINEC score for Vibrio-related skin infections, however, has been called into question.³⁷ In Vibrio vulnificus-infected patients, a LRINEC score ≥2 and hemorrhagic bullous or blistering skin lesions were significant predictors of NSTI.³⁷ In a systematic
review, the most common risk factors for developing necrotizing fasciitis included IV drug abuse, smoking, trauma, and diabetes.³⁸