Part III: Management of Hand Infections and Deep Tissue Injuries
Author: Gary D. Hals, MD, PhD, Attending Physician, Department of Emergency Medicine, Palmetto Richland Memorial Hospital, Columbia, SC.
Peer Reviewers: Alasdair K.T. Conn, MD, Chief of Emergency Services, Massachusetts General Hospital; Andrew Perron, MD, FACEP, Assistant Professor of Emergency Medicine, Associate Residency Director, Department of Emergency Medicine, University of Virginia, Charlottesville.
Hand infections are common and, if not treated properly early in their course, can lead to devastating outcomes. Bacterial sources are the most frequent cause, but viral and fungal pathogens also can prompt emergency department (ED) visits. Occult foreign bodies always should be in the diagnosis for hand infection, since healing only will be achieved after removal of the nidus of infection. Several anatomic features of the hand lend it to rapid and deep spread of infection compared to other body parts. Infection can spread rapidly from the distal finger along the flexor tendon sheath to the palm. The fibrous connections of the fingertips and palm leave little room for swelling and can aid in proximal spread, or force the infection to the dorsal surface of the hand. Often, the exam of the palm will be benign enough that it can be overlooked as the source of the infection. Further, as dorsal infections generally are better tolerated than palm infections, this presentation can mislead the ED physician as to the true nature of the problem. Even if accurately diagnosed and treated, any hand infection has the potential to result in loss of hand function from tendon scarring, nerve, and tissue damage.
To address these issues, the final part of this three-part series focuses on diagnosis and management of hand infections and deep tissue injuries. Guidelines for patient evaluation, surgical consultation, and initial antibiotic therapy, when indicated, are provided.—The Editor
Table 1 provides a summary of warning signs to look for when treating a patient in the ED for a hand infection. The following sections will discuss diagnosis and treatment of common hand infections and also briefly will discuss disorders that may mimic infection upon presentation.
Hand Infection Mimics. A few of the more common disorders that result in inflammation of the hand without infection bear discussion because they can be very misleading and result in misdiagnosis in the ED. Gout is an inflammatory arthritis that occurs most commonly in the lower extremity (podagra) but can appear in the hand without previous history.1 Urate crystals precipitate out in a joint, leading to swelling, erythema, and pain with motion. Joint aspiration yielding urate crystals is diagnostic, but consultation is indicated when no fluid is obtained or the diagnosis is uncertain. Herpetic whitlow is perhaps the most common mimic of hand infection. Although it is a true hand infection, the source is viral and not bacterial. Herpetic whitlows easily can be confused with paronychias and felons. These infections are common in health care workers and small children. Early in the infection, pain out of proportion to physical findings can be seen. The vesicle fluid initially will be clear, and surrounding erythema will appear around the base of the fingernail. As vesicles coalesce, the fluid will become cloudy, and it is this stage that most easily is confused with a bacterial infection. The distinction is important because drainage of a herpetic whitlow is contraindicated; it may cause secondary bacterial infection and viral dissemination. It is important to recognize these when a patient presents to the ED, because they are treated with dry dressings and oral antiviral drugs, not with incision and drainage. Lesions will take 12-14 days to heal.
Acute Paronychia. A paronychia is a bacterial infection at the margin of the nail bed. Together with felons (see next section) they are the most common hand infections, making up one-third of cases.2 It is important to distinguish a paronychia from a felon (infection of the pulp space of the fingertip), as untreated felons can lead to osteomyelitis and even amputation. A paronychia begins when bacteria enter the skin via small tears in the eponychial fold (i.e., hangnail). In contrast to a felon, the infection is confined to tissue adjacent to the nail and does not involve the finger pad. A paronychia can spread to produce a felon as well (and vice versa), but the important point when diagnosing a paronychia is to examine the finger for the presence of a felon.
The most common offending organism is Staphylococcus aureus. Nail biting can increase the incidence of these infections and introduce oral flora (anaerobes) as well. Paronychias begin with cellulitis, and some will resolve spontaneously. If they progress, the patient typically presents with a history of several days of pain and swelling around the nail and often has a small collection of pus visible on exam. Treatment consists of local incision and drainage of the pus followed by warm soaks to encourage drainage. (See Figure 1.) If more advanced with pus spread underneath the nail, then the involved section of the nail should be elevated and at times removed.2 Gauze should be placed between the eponychium and the nail bed to ensure adequate drainage and increase likelihood of subsequent normal nail growth. Antibiotics are not needed routinely for simple acute paronychias, but some authors still recommend their routine use.3 Most paronychias will resolve in 5-10 days. Chronic paronychias may benefit from antibiotic or antifungal therapy.
Felon. A felon is an abscess of the fingertip pulp, and commonly arises from a puncture wound of the fingertip. Untreated paronychia also can spread to involve the finger pulp. These are serious infections that can result in necrosis of the fingertip, osteomyelitis, and ultimately amputation if not treated properly. Continued accumulation of pus will increase pressure in the fingertip as the fibrous septae of the pulp initially will prevent spread of the infection proximally. Pus can track onto the dorsal surface and create a paronychia as well as spreading into the bone. Delay in treatment of a felon can cause spread of the infection proximally to produce flexor tenosynovitis and/or a septic distal interphalangeal (DIP) joint. However, the inflammation and pressure in the fingertip make a felon very painful, and most patients present after only a few days of symptoms. As with the paronychia, S. aureus is the most frequent bacterial cause.
There is no disputing the fact that a felon is an abscess in need of incision and drainage, but considerable controversy remains concerning the type of incision to make. One group of hand surgeons suggests making a fish-mouth or J-shaped incision along the side of the finger.4 This group chooses to avoid incision directly onto the finger pad because scars on the finger pad can be very sensitive and may cause long-term problems (i.e., neuromas). In addition, they believe that all sepatal spaces in the fingertip must be opened to ensure proper drainage and healing. Another group suggests incisions be made where the abscess points, typically at the finger pad.2,5 (See Figure 2A.) They recommend avoiding fish-mouth incision as it may compromise blood supply of the finger pad and produce painful scarring. One author further suggests that longitudinal volar incisions be used, as horizontal ones inadvertently may transect the terminal branches of digital nerves.2 As all patients being treated for felons should have close follow-up with a hand surgeon, perhaps the best advice is to consult the hand surgeon before the felon is drained to allow his input on the type of incision to be used. Green’s text on hand surgery reports that volar incisions are best done when the abscess already is draining there, and that longitudinal incisions produce the best result. No controversy on antibiotic use exists. In general, a 10- to 14-day course of a first-generation cephalosporin is recommended. Indications for admission include signs of systemic involvement (i.e., fever, vomiting), failure of outpatient treatment, immunosuppression (i.e., diabetes, AIDS, end-stage renal disease [ESRD], etc.), or lack of reliable follow-up (i.e., homeless, intoxicated, etc.).
Flexor Tenosynovitis. Flexor tenosynovitis is a closed-space infection of the flexor tendon sheath and/or the radial or ulnar bursae. Extensor tendons do not develop classic tenosynovitis because they lack the unique tendon sheath found on flexor tendons. Figure 3 shows the anatomy of the flexor tendon sheaths and the radial and ulnar bursae. In 80% of people, these two bursae communicate in the palm area and potentially can permit spread of infection from one digit to the others.
These infections usually begin from penetrating trauma with subsequent inoculation of the flexor tendon sheath, but rarely can arise from hematogenous spread. Subsequently, skin flora (S. aureus) are the most common bacteria causing tenosynovitis. Nutrients in the tendon sheath naturally help maintain the tendons, and unfortunately also feed bacteria deposited there. Pressure in the tendon sheath quickly rises, inhibiting access of the immune system to fight the infection. Increased pressure also reduces blood flow to the area, optimizing conditions for bacterial growth and increasing risk of tendon damage. It is this pressure that also is responsible for the common presenting signs of flexor tenosynovitis: Kanavel’s cardinal signs.
In 1925 Kanavel described four cardinal signs of flexor tenosynovitis: pain on passive extension, tenderness along the course of the flexor tendon, symmetric (fusiform) swelling of the involved finger, and a flexed position of the finger.6 Pain on passive extension is thought to be the most accurate of the signs, and tenderness along the tendon course is thought to be the most helpful in distinguishing tenosynovitis from mimics (i.e., gout, herpetic whitlow). Since complications of tenosynovitis can be so devastating, an experienced hand surgeon should be consulted to make the final call when the diagnosis is in question. Sometimes surgical intervention is necessary to ensure the diagnosis.
Tenosynovitis tends to be a rapidly progressing infection, with most patients presenting within just 48 hours after an injury.7 All patients with flexor tenosynovitis require urgent consultation with a hand surgeon and admission to the hospital. Although, in theory, an early case could be treated with antibiotics alone, such early cases rarely are encountered, and the surgical hand literature still recommends admission to allow close observation of progress of the infection.8 Most patients can be treated by the hand surgeon with drainage and irrigation of the flexor tendon sheath and IV antibiotics. Choice of antibiotic usually is guided by wound culture, with ceftriaxone given empirically until results are available.9
Septic Arthritis. Although not common, any joint in the hand can become infected. It must be recognized and treated early to avoid loss of use of the joint. It can occur from hematogenous spreading, but this mode of infection occurs less commonly in the hand than in other joints of the body (i.e., hip, knee, etc.). When seen, it is more often in immunocompromised patients, especially with diabetes, alcoholism, intravenous drug abuse (IVDA), and ESRD. Septic arthritis of the hand usually occurs as a complication of trauma or from direct spread of an adjacent infection, such as tenosynovitis. A septic metacarpophalangeal (MP) joint may result from an untreated fight bite (human bite wound). Patients usually will present complaining of severe pain in the joint and may have a puncture wound overlying the joint. Cat bites directly over a finger joint are fairly common and have a high risk for joint involvement. The joint will be held in mild flexion, the position that maximizes the volume of the joint and minimizes the pain. Any passive motion will be very painful. Skin over the joint typically is red and very swollen. The white blood cell (WBC) count is unreliable because it is elevated in fewer than 50% of patients,10 and the erythrocyte sedimentation rate (ESR) and C-reactive protein levels may be more useful. An x-ray should be obtained because bony destruction may be evident. The diagnosis can be made only via arthrocentesis, although very little fluid may be present in some hand joints. If fluid is obtained, WBC count greater than 50,000/mm3 or neutrophil count greater than 75% of total WBC count confirms the diagnosis.10 However, when the patient presents early in the course of infection, lower WBC counts may be obtained. The ED physician should maintain a high index of suspicion and not be falsely reassured by low WBC counts in the face of a good clinical presentation. Gram’s stain of the fluid also can be very helpful in identifying infected joints. Any patient who may have a septic joint should have an emergent consultation by the hand surgeon. A non-diagnostic arthrocentesis by the ED physician will not rule it out. Treatment ordinarily is done with open irrigation of the joint and IV antibiotics as an inpatient. Some hand surgeons may use closed irrigation of the joint instead.11 Regardless of the method of treatment, patients who are treated early (within 24 hours of symptom onset) have a much better prognosis for return of joint function.
Hand Cellulitis. Simple cellulitis of the hand is not as common as most other types of hand infections discussed in this section. When treating simple cellulitis of the hand, the ED physician must remain vigilant for other, more serious conditions masquerading as a simple problem. Cellulitis in the finger should prompt the physician to rule out flexor tenosynovitis. Cellulitis over a joint requires ruling out a septic joint as the underlying condition. Hand cellulitis can occur without deep abscess formation, but abscess presence must be ruled out, and consultation with the hand surgeon often is necessary to do so. Some patients may present with complaint of a "hand infection," only to be found with an untreated laceration. Any infected laceration associated with an open fracture, tendon injury, or other deep structure involvement requires consultation for admission. At times, infection can arise spontaneously from a foreign body embedded in the hand. This can occur acutely or many months after the injury. History of a possible foreign body or remote trauma to the area should alert the physician to evaluate for a retained foreign body. (See section on foreign bodies.) A limited, truly simple cellulitis of the hand can be treated with oral cephalosporins (cephalexin 500 mg QID or 25-50 mg/kg/day in children). Any "simple cellulitis" that involves a large area of the skin or occurs in an immunocompromised patient (i.e., HIV, ESRD, diabetes, etc.) should be treated as an inpatient with IV antibiotics.
Necrotizing Fasciitis. One may not normally picture the hand when thinking of necrotizing fasciitis, but the upper extremity is one of the most common sites for this infection. In Giuliano’s paper that originally divided necrotizing fasciitis into two types based on the bacteria involved, 68% of the cases were in the upper extremity.12 He divided these infections into type I (polymicrobial) and type II (group A streptococcal) infections. Type I infections are more common on the trunk, but type II infections occur more frequently on the upper and lower extremity. Group A streptococcal infections also are referred to as "flesh-eating bacteria" by the media. This infection more often is seen in previously healthy patients, and can progress rapidly from symptom onset. Patients who abuse injection drugs are at particularly high risk for soft-tissue infections, including necrotizing fasciitis. One series of 33 cases of necrotizing fasciitis of the upper extremity found that all but one were in patients with history of IVDA.13 More recent reviews found that 55% of patients at University of CaliforniaDavis Medical Center with necrotizing fasciitis resulted from complications of IVDA.14 A series of unexplained deaths of IVDAs in 2000 in Europe also was blamed on undiagnosed necrotizing fasciitis.15 Unfortunately, soft-tissue infections from IVDA are very common in some communities. Data from 1996 to 2000 at San Francisco General Hospital revealed that soft-tissue infection from IVDA patients was among the most common admitting diagnoses overall.16 San Francisco General Hospital lost $18 million in unreimbursed dollars from IVDA skin abscesses in 1999 alone.
The rapid course seen in necrotizing fasciitis accounts for much of the mortality (up to 70% in some series) associated with this infection. Early in the course, the patient may present with very minor findings on the surface of the skin but severe complaints of pain. As the infection progresses, erythema of the skin appears along with non-pitting, woody edema extending out from the erythema. This is due to the pathological process at work. The infection spreads along fascial planes under the skin, and the skin itself may not show signs of the infection until the fascia under it is heavily involved. Once bacteria enter the body, exotoxins produce liquefactive necrosis of subcutaneous fat and fascia, along with thrombosis of end arteries (obliterative endarteritis). The anaerobic conditions produced by these two processes promote very rapid bacterial growth followed by further necrosis and ischemia. Fascial destruction rates as high as 2-3 cm per hour have been quoted by some authors,17 underscoring the importance of avoiding delayed diagnosis and treatment. If the cycle is not interrupted by surgical intervention and antibiotics, the patient stands little chance. In a review of 12 cases of upper extremity necrotizing fasciitis, an average of three debridements were necessary.18 Two patients underwent amputation at the shoulder. In another review of 33 patients with similar infections, seven patients had complications of shock, resulting in three deaths and two amputations.13 All of the severe complications were found in patients who underwent debridement more than 24 hours after presentation.
Unfortunately, diagnosis of necrotizing fasciitis is fraught with pitfalls. One study found 41% of patients were discharged from the ED on initial presentation after only being treated for an abscess.18 Infection spreads so quickly, that the patient’s WBC and temperature may not be elevated at presentation. One must maintain a high index of suspicion in the proper clinical setting to detect these cases. Any patient who presents with pain out of proportion to clinical findings with a skin infection should be considered to possibly have necrotizing fasciitis. History of IVDA and/or HIV infection from IVDA are considered major risk factors. Besides initial screening bloodwork (which should not be considered diagnostic), plain films may be of use if subcutaneous gas is found. Gas usually is due to infection with Clostridial organisms, which is less common than Streptococcal infections.19 Clostridial infections more often are associated with soil contamination of wound, as seen in farm injuries. Regardless of the bacterial cause, patients suspected of having necrotizing fasciitis should have emergent consultation by a hand surgeon. Successful treatment in based on rapid surgical debridement, fluid resuscitation, and antibiotic therapy, in order of importance. Diabetic hand infections are less common than foot infections, but when they present, they can be confused with necrotizing fasciitis. Usually they will have a much longer course (days) compared to necrotizing fasciitis (hours).
Deep Space Infections. There are five potential spaces located deep inside the hand that represent possible sites for abscess formation: thenar space, midpalmar space, hypothenar space, dorsal subaponeurotic space, and the interdigital web space. (See Figure 4.) Although these infections are not very common, they do make up 5-15% of all hand infections.20 These potential spaces represent closed compartments where penetrating trauma, progression of flexor tenosynovitis, or hematogenous spread can deposit bacteria with rapid development of dangerous abscesses. The physical exam in patients with deep space infections allows the physician to diagnose which space is involved, as infection in each space produces a characteristic set of physical findings. As the skin on the palm is less flexible than the dorsum, the majority of the swelling and purulence may be more noticeable on the dorsum of the hand. One should be aware that the source of the infection lies in the palm in these patients and incorrect drainage will not resolve the infection. A thenar space abscess presents with dramatic swelling and pain at the base of the thumb. Increased pressure in this space pushes the thumb to an abducted and flexed position to increase the volume of the thenar space. Passive or active thumb movement will be limited dramatically due to pain. The midpalmar space is located at the base of the ring and little fingers. Abscess formation here produces pain and swelling in the central palm; loss of the normal palm concavity is diagnostic. The dorsal subaponeurotic space is located beneath the extensor tendons and above the metacarpals. The dorsal hand is swollen and painful, and attempts at finger extension are limited by pain. The clinical presentation of this abscess often is very similar to dorsal cellulitis or a simple subcutaneous abscess, so that diagnosis often is not certain until drainage by the hand surgeon. An interdigital or web space infection often results from infection of a skin fissure, palmar callus, or blister. Although the infection begins on the palm, the fibrous connections on the palm prevent significant spread there and result in spread to the dorsal web space. This abscess is referred to as collar-button abscess because of its hour-glass shape (resembling collar buttons of dress shirts in the early 1900s). Adequate drainage of this abscess often can be problematic, as incisions must be made both on the dorsal and palmar skin to ensure drainage. Incisions in the webspace itself are not done; scarring here will prevent adequate finger abduction.20 A hypothenar abscess is rare, but can occur with swelling and pain at the base of the fifth finger.
Treatment of deep space hand infections involves adequate incision and drainage by an experienced hand surgeon and is beyond the purview of the emergency physician. It is important to properly diagnose these abscesses so that the consultant is aware of the true nature of the problem, and the emergency physician is not fooled by thinking he is dealing with a superficial infection. IV antibiotics (first-generation cephalosporin) can be started in the ED while waiting for the consultant to arrive.
Pediatric Hand Problems
Up to 20% of visits by pediatric patients to the ED will be for the results of trauma,21 and a large percentage of these will involve upper extremity and hand injuries. To a large degree, treatment of pediatric hand problems mirrors that of adults, but there are some significant differences. From the outset, many deductions can be made from clinical exam in adults, but children are less cooperative with physical exams and may not allow useful deductions to be made easily. For example, very young children lack the necessary motor control for adequate tendon exam. When discussing fractures, the primary issue is the presence of growth plates. Whereas adults tend to sprain ligaments more often, children tend to fracture around growth plates. These fractures are classified by the Salter-Harris system. While children tend to heal better than adults, growth plate injuries bring a much higher risk for permanent damage by asymmetric growth than comparable adult fractures. Different age groups are more likely to break different hand bones. One study of hand fractures in children found that distal tuft fractures were most common in the 0- to 8-year-olds, while in the 13- to 16-year-old group metacarpal fractures were most frequent.22 Mallet equivalent fractures in children also are treated differently than in adults. While the fingers look the same clinically, these injuries are due to an open displaced fracture in children, whereas in adults they are due to a closed tendon disruption. Thus, the adult injury can be treated as an outpatient while the child’s injury will need consultation in the ED.
Plain films can be much more difficult to interpret in young children due to the presence of growth plates. Growing bones, especially in the wrist, lack enough calcium even to show up on x-ray, making interpretation of injuries there very difficult. Comparison films can be of use when questionable findings arise. Growing bones often remodel in younger children, but when treating hand injuries the physician should be aware that angulation in the A/P direction will not remodel, whereas angulation in the lateral plane will. Soft tissue healing in children often will be superior to that of an adult, but it should not be counted on in every patient. Children are, by nature, less compliant than adults, and it is difficult to elevate their hands, for example. Vessels are much smaller and more difficult to reattach in the case of amputations as well. The physician also must remember that a child’s compliance will be dependent on the parent’s, and social considerations must be taken in to account when discharging patients for follow-up. Likewise, the physician should follow his instincts concerning abuse when treating hand injuries. Maintain a low threshold to consult social services when clinical findings do not match the history given. The physician should trust his "gut instincts" when considering child abuse. Finally, animal bites may be more common in children than in adults. Up to 50% of dog-bite injuries are in children younger than 12 years old, and half of these are on the hand or forearm.23 Further, one-third of the fatalities annually from dog bites are in children.23
Bites Injuries: Human and Animal. Regardless of the species involved, bite injuries present a risk for infection as well as for tissue damage. Teeth can cause puncture wounds that penetrate joints, crush tendons or nerves, and lacerate skin. In general, each wound must be evaluated thoroughly to rule out injury to deeper structures. Remember to put fingers and tendons through their range of motion while looking in the wound to be sure no "hidden" injuries are present. The common finding in fight bites is that the MP joint and/or extensor tendon are damaged while the fist is clenched, and then the tissue shifts over these injuries when the hand is unclenched and examined in the ED. Only when shifting the hand back into the position at time of injury will the deeper damage be seen. Further, bacteria will be deposited deep in the hand and will be in an anaerobic environment, encouraging rapid bacterial growth.
Although other animals account for far more bites in terms of numbers, the human bite wound, or fight bite, is one of the most dangerous hand injuries of all. Injury to bone, cartilage, or tendon will be seen in 75% of these patients. It is important to state at the outset that the standard of care in the literature is for any infected fight bite to be admitted to the hand consultant for open irrigation and IV antibiotics.24 One randomized study found that of patients who presented in fewer than 24 hours, had no signs of infection, and had no deep structure injury, 50% still became infected.25 The poor compliance of this patient population is evidenced by the high incidence of late presentation, against medical advice (AMA) self-discharge, and lack of return for follow-up care.26 Another review of 100 human bite injuries to the hand found that the amputation rate was 18%, and amputation was much more likely when infections became established in the bone or tendon sheath.27 These patients should not be sent home by the ED physician.
Although this injury has likely been a problem as long as people have been fist-fighting, it was not described in the literature until 1911.28 Large numbers of bacteria are present in the human mouth, with an estimated 100 million organisms per milliliter of saliva and up to 42 different species of bacteria.29 The most common organisms in hand infections from human bites overall are S. aureus and Streptococcus, but Eikenella corodens is seen in up to 30% of fight bites.29 This may be due to the fact that Eikenella is more prevalent on tooth scrapings than in saliva, and that the hand is inoculated by the fight bite mechanism. The hand can be exposed to infection from human saliva by fingernail biting and sucking a bleeding wound, or by direct biting of the hand with traumatic amputation. While the mechanism of these infections tends to be easily provided in the history, people are not so forthcoming with accurate information on fight bites. Some individuals will present early, fewer than 24 hours after the incident, while others will delay treatment and present more than 24 hours after. This latter time is used by many authors as the incidence of complications, and type of treatment indicated will differ between the two groups.30 The early patients may be as secretive about the true mechanism of injury as those presenting late, so the ED physician must consider every laceration or puncture over the MP joint area as a possible fight bite. Plain films will help when they show a metacarpal head fracture, small bony defect, or a retained tooth fragment, but films more often are normal. Any patient with suspicion of a fight bite should have the wound explored under sterile conditions to rule out extensor tendon or MP joint capsule injury. The hand must be examined in a clenched-fist position to properly line up these structures under the wound or the physician easily may miss a deeper injury. The early patient may not have signs of infection yet but should be treated with amoxicillin/clavulanate 500 mg PO q 8 hours, or clindamycin 300 mg PO q 6 hours plus ciprofloxacin 500 mg PO BID (for penicillin-allergic patients), and have the case discussed with the hand consultant. The best scenario is for the patient to be seen by the hand consultant while in the ED, but referral to his office may be more realistic. Specific and thorough discharge instructions about the importance of follow-up, risk of non-compliance, etc., should be given to the patient and documented in the chart. Late-presenting patients (more than 24 hours) usually are brought to the ED by increasing pain from an established infection, and are in the highest risk category for complications. Documented complications vary from joint stiffness and arthritis to osteomyelitis, amputation, sepsis, and even death.31 Once scarring has begun to the volar plate, collateral ligaments, and tendons, the chances for retaining finger function are minimal. These patients should be referred for admission and surgical debridement by the hand surgeon.
Dogs account for 80-90% of the animal bites in the United States,23,32 cats for only 5-10%, and up to 10% are caused by other animals. In other terms, animal bites cause around 300,000 ED visits, or 1% of all visits;33 10,000 admissions; and 20 deaths annually.32 Up to 50% of all Americans will suffer a mammalian bite wound in their lifetimes.34 Dogs tend to cause more tissue damage, while cats produce puncture wounds rather than avulsions. Thus, cat bites have higher infection rates, as they are more difficult to irrigate. The hand is involved more often, as it frequently is closer to the animal than other parts of the body. One study found that 50% of infected dog bites and 63% of infected cat bites occurred on the hand.35 The most common bacteria isolated from these injuries is Pasteurella multocida, found in 50% of infected dog bites and 75% of infected cat bites.35 Streptococcus and Staphylococcus are the next most common organisms. Infection moves quickly in animal bites, as the average time for onset of symptoms was only 12 hours for cat bites and 24 hours for dog bites.32 Copious irrigation with normal saline is the first line of defense for infection. Prophylactic antibiotics are indicated, and meta-analysis of dog bite wounds shows that prophylactic antibiotics do reduce the infection rate.36 The infection rate for controls was 16%, and was cut in half with antibiotic use. While most authors recommend prophylactic antibiotics, another meta-analysis found no evidence of reduction in infection after dog or cat bites, but did find a difference for human bites.33 Primary closure of hand bite wounds generally is not recommended, although in one study the infection rate for primary closure of mammalian bites (to all parts of the body) was only 6%.37 This may be acceptable when cosmetic issues are present (facial injuries), but unless tendon or bone is exposed, the physician only should loosely pull together hand bite wounds when deep structures are exposed. With such injuries, most ED physicians will defer even primary care of these wounds to the hand consultant.
Compartment Syndrome. Compartment syndrome occurs when the pressure in a closed fascial space exceeds 20 mmHg to 30 mmHg. At this cutoff, perfusion is reduced, causing muscle/nerve ischemia, and a vicious cycle of further swelling producing worsening ischemia, is developed. This culminates in necrosis and potential amputation that can follow unless surgical treatment intervenes to break the cycle. Patients can present with compartment syndrome of the hand and/or forearm from a multitude of causes. Some examples include: supracondylar fractures, closed fractures with multiple fracture sites, tight-fitting casts, high-pressure injection, IV infiltration, venomous snake bites, rhabomyolysis from drugs, seizures, prolonged immobility, crush injury, or burns. The most common cause of compartment syndrome in the hand is arterial line malfunction or IV line infiltration. Misdiagnosing compartment syndrome can be disastrous, as it often will result in Volkmann’s ischemic contracture, a permanent and virtually untreatable flexion of the wrist and hand.38 Early detection of compartment syndrome is key in treatment as after 4-8 hours of delay permanent damage is more likely to occur. Often patients at risk for this be seen in the ED, have their other problems addressed, and then be admitted to the hospital. It is important to recognize the potential risk of compartment syndrome and continue to monitor for its development while these patients remain in the ED. As many of the clinical findings rely on the patient’s feedback, obtunded patients represent a unique risk for detecting compartment syndrome. One study of 17 patients with compartment syndrome of the hand found that 88% were obtunded when the compartment syndrome was diagnosed.39 All of the patients who subsequently went on to amputation or permanent loss of hand function were in the obtunded group.
Clinical findings in compartment syndrome can vary, but pain on passive motion often is cited as the most reliable finding. Pain on passive stretch of the muscles involved is very suspicious, and always should raise concern for the diagnosis. Ultimately, the patient will not be able to move the affected digits. As damage progresses, patients will present with dramatic swelling that will continue with time. The hand is described as "rock hard," and it follows that if swelling of a compartment is reducing blood flow, that the swelling should produce a very tight compartment. Pallor and pulselessness often are described, but actually are late findings. Dramatic loss of sensation also is a late finding, but early on decreased sensibility to light touch or pinprick can be detected. The fingers will be held with slight flexion at the PIP and DIP joints, and with the MP joints extended. Any patient in whom compartment syndrome is suspected should have immediate consultation by a hand surgeon or transfer arranged to a facility where this can be performed. Compartment pressures need to be monitored, and there are 10 individual compartments in the hand.40 Many texts quote pressures of 20 mmHg to 40 mmHg, with most authors using 30 mmHg as the benchmark. Specifically for the hand, however, some suggest using reduced numbers than those used in evaluation of lower extremities, and recommend considering surgery at 15 mmHg to 20 mmHg.39 Still others suggest that no specific pressure should be used and pressures should be interpreted in the context of each patient’s clinical exam.41 No debate exists on treatment for compartment syndrome, and immediate fasciotomy of each affected fascial compartment is indicated as soon as the diagnosis is made. While evaluating a patient for compartment syndrome, one may be tempted to elevate the limb to reduce swelling. This should not be done, as elevation above the level of the heart only will serve to reduce the ateriovenous gradient, which actually decreases tissue blood flow and further increases edema.
High-Pressure Injection. These injuries are a good example of ED complaints that can be occult in initial presentation. They present with very minor external trauma, but often have extensive damage beneath the skin similar to high-voltage injuries, fight bites, necrotizing fasciitis, etc. Although early reports exist of high-pressure injection injuries in 1937,42 it was not until the more recent decades that these instruments became more common in the workplace and home thus increasing the incidence of injury. The pressures produced can range from 1000 psi to more than 10,000 psi, and only 100 psi is needed to easily break the skin.43 To put these pressures in perspective, the kinetic energy involved is similar to a 450-lb weight falling 25 cm.44 Most accidents occur on the non-dominant hand as the worker tries to clear an obstructed nozzle. The index finger is the most common site of injury, followed by the middle finger and palm.
Important factors in these injuries include: the type of material injected, the amount injected, the location of the injury, and the pressure involved. Many liquids are used by these machines, and the most common ones seen in injection injuries, in order of frequency, include grease, paint hydraulic fluid, diesel fluid, paint thinner, molding plastic, paraffin, and cement.45 Unfortunately, oil-based products (i.e., paint, paint thinner, etc.) cause the most tissue damage by producing a large inflammatory response resulting in fibrosis. Injection of oil-based paint is one of the leading risk factors for amputation, cited as a cause in 70% of amputations from high-pressure injections.46 The pressure is directly related to the amount of material injected, and produces harm by direct compression of vessels, vasospasm, and larger inflammatory response. The amount of material injected also depends on the location of the injury. The flexor tendon system is entered more easily at the DIP and PIP flexion creases. As the tendon sheaths of the index, middle, and ring fingers end around the MP joint, they help to limit proximal spread of material. The sheaths for thumb and little finger, however, connect to the radial and ulnar bursae and permit spread into the forearm. This will increase risk for compartment syndrome and ulnar/median nerve entrapment at the carpal tunnel. Amputation rates vary and may be as high as 100%, as described with injection pressures over 7000 psi.47 Delayed treatment or presentation more than 10 hours after the injury also is associated with higher amputation rates.
When the patient presents early, there will be minimal complaints and often only a small laceration or pinhole at the site of injury. It is important that the ED physician recognize the significance of the injury, as within the next 4-6 hours the patient will experience severe pain and swelling that may progress to tissue ischemia and necrosis. If left untreated, skin breakdown may occur over areas of accumulated material, which opens the window for bacterial superinfection. Once a patient is identified in the ED, plain films should be taken as they may help identify the extent of proximal spread of the injected material. The hand surgeon should be consulted immediately, as these patients will require debridement and admission for observation and further treatment. Failure to properly refer these patients is cited as a cause for negligence claims.48 It is important to remember not to use digital blocks when treating these patient’s pain, as this may worsen digital ischemia. One author suggests early amputation in any digit that presents with limited perfusion.49 However, others, using an open wound treatment, report a high salvage rate of 84%, with 64% of patients returning to normal function and 92% returning to their previous job.50
Foreign Bodies. Retained foreign bodies are relatively common in the hand and are cited among the more common causes for litigation against ED physicians. Any hand wound treated in the ED should be considered as possibly harboring a foreign body. The most common foreign bodies found in the hand include wood, glass, and metal.51 As some will be radio-opaque, one should use clinical clues to help identify these cases. Patients should be questioned specifically about a foreign body sensation, and the area around the wound palpated. Any mass or wound that has failed to heal in a normal time frame is highly suspect. Local exploration with good anesthesia in a bloodless field is probably the best method of detection. With a strong suspicion but no findings on exam or with x-ray, some have recommended use of ultrasound to help localize retained foreign bodies. One study found that ultrasound could detect wooden splinters of only 2.5 mm in length with a high degree of accuracy.52 The bottom line is that if a patient feels that a foreign body is present, it most likely is even if the physician cannot find it. Patients with a suspected or partially removed foreign should be counseled that a foreign body(ies) still may be retained in the wound. Follow-up should be arranged with a hand surgeon, prophylactic antibiotics given, and the dangers of non-compliance with follow-up stressed and documented in the chart. Even patients where "all" of the foreign body is retrieved should be informed that there still may be something left behind, and given instructions on what to look for next.
1. Louis DS, Jebson PJ. Mimickers of hand infection. Hand Clin 1998;14:519-530.
2. Jebson PJ. Infections of the fingertip: Paronychias and felons. Hand Clin 1998;14:547-556.
3. Rockwell PG. Acute and chronic paronychia. Am Fam Physician 2001;63;1113-1116.
4. Lester B. The acute hand. Chapter 19—Amputation. Appleton & Lange, 1999;287-288.
5. Kilgore ES Jr, Brown LG, Newmeyer WL, et al. Treatment of felons. Am J Surg 1975;130;194-198.
6. Kanavel AB. The treatment of acute suppurative tenosynovitis—discussion of the technique. In: Infections of the Hand, 5th Ed. Philadelphia, Lea & Febiger 1925:225-226.
7. Boles SD, Schmidt CC. Pyogenic flexor tenosynovitis. Hand Clin 1998;14:567-578.
8. Neviaser RJ. Tenosynovitis. Hand Clin 1989;5:525-531.
9. Levy CS. Treating infections of the hand: Identifying the organism and choosing the antibiotic. In: Green WB, ed. Instructional course lectures, Parkridge AAOS 1990:vol 59.
10. Murray PM. Septic arthritis of the hand and wrist. Hand Clin 1998;14:579-588.
11. Harris PA, Nanchahal J. Closed continuous irrigation in the treatment of hand infections. J Hand Surg 1999;24:328-333.
12. Giuliano A, Lewis F, Handley K, et al. Bacteriology of necrotizing fasciitis. Am J Surg 1977;134:52-57.
13. Schecter W, Meyer A, Schecter G, et al. Necrotizing fasciitis of the upper extremity. J Hand Surg 1982;7:15-20.
14. Chen JL, Fullerton KE, Flynn NM. Necrotizing fasciitis associated with injection drug use. Clin Infect Dis 2001;33:6-15.
15. Williamson N, Archibald C, Van Vliet JS. Unexplained deaths among injection drug users: A case of probable Clostridium myonecrosis. CMAJ 2001;165:609-611.
16. Soft tissue infections among injection drug users—San Francisco, California, 1996-2000. MMWR Morb Mortal Wkly Rep 2001;50: 381-384.
17. Laucks SS 2nd. Fournier’s gangrene. Surg Clin North Am 1994;74: 1339-1352.
18. Gonzalez MH, Kay T, Weinzweing N, et al. Necrotizing fasciitis of the upper extremity. J Hand Surg 1996;21:689-692.
19. Gonzalez MH. Necrotizing fasciitis and gangrene of the upper extremity. Hand Clin 1998;14:635-646.
20. Jebson PL. Deep subfascial space infections. Hand Clin 1998;14: 557-566.
21. Singer JI. Pediatric Injuries. In: Hart RG, Uehara DT, Wagner MJ, eds. Emergency and Primary Care of the Hand. Irving, TX, The American College of Emergency Physicians 2001:259-270.
22. Rajesh A, Basu AK, Vaidhyanath R, et al. Hand fractures: A study of their site and type in childhood. Clin Radiol 2001;56:667-669.
23. Synder CC. Animal bite infections of the hand. Hand Clin 1998;14: 691-712.
24. Kelly IP, Cunney RJ, Smyth EG, et al. The management of human bite injuries of the hand. Injury 1996;27:481-484.
25. Zubowicz VN, Gravier M. Management of early human bites of the hand: A prospective randomized study. Plast Reconstr Surg 1991; 88:111-114.
26. Tonta K, Kimble FW. Human bites of the hand: The Tasmanian experience. Aust N Z J Surg 2001;71:467-471.
27. Mennen U, Howells CJ. Human fight-bite injuries of the hand. A study of 100 cases within 18 months. J Hand Surg 1991;16: 431-435.
28. Peters WH. Hand infection apparently due to Bacillius fusiformus. J Infect Dis 1911;8:455-461.
29. Shields C, Patzakis MJ, Meyers MH, et al. Hand infections secondary to human bites. J Trauma 1975;15:235-236.
30. Mann RJ, Hoffeld TA, Farmer CB. Human bites of the hand: Twenty years of experience. J Hand Surg 1985;2:97-104.
31. Long WT, Filler BC, Cox E II, et al. Toxic shock syndrome after a human bite to the hand. J Hand Surg 1988;13:957-959.
32. Wolanyk DE. Infections. In: Hart RG, Uehara DT, Wagner MJ, eds. Emergency and Primary Care of the Hand. Irving, TX, The American College of Emergency Physicians, 2001:219-246.
33. Medeiros I, Saconato H. Antibiotic prophylaxis for mammalian bites. Cochrane Database Syst Rev 2001;(2):CD001738.
34. Smith PF, Meadowcroft AM, May DB. Treating mammalian bite wounds. J Clin Pharm Ther 2000;25:85-99.
35. Talan DA, Citron DM, Abrahamian FM, et al. Bacteriologic analysis of infected dog and cat bites. Emergency Medicine Animal Bite Infection Study Group. N Engl J Med 1999;340:85-92.
36. Cummings P. Antibiotics to prevent infection in patients with dog bite wounds: A meta-analysis of randomized trials. Ann Emerg Med 1994;23:535-540.
37. Chen E, Hornig S, Shepherd SM, et al. Primary closure of mammalian bites. Acad Emerg Med 2000;7:157-161.
38. Botte MJ, Keenan MA, Gelberman RH. Volkmann’s ischemic contracture of the upper extremity. Hand Clin 1998;14:483-497.
39. Ouellette EA. Compartment syndromes of the hand. J Bone Joint Surg Am 1996;78:1515-1522.
40. Antosia RE, Lyn E. The hand. In: Rosen P, Barkin R, Hockberger R, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th ed. Mosby; 1998: 662-668.
41. Ortiz JA Jr, Berger RA. Compartment syndrome of the hand and wrist. Hand Clin 1998;14:405-418.
42. Rees CE. Penetration of tissue by fuel oil under high pressure from diesel engine. JAMA 1937;112:907-908.
43. Proust AF, Stubitsch BT. Unique injuries. In: Hart RG, Uehara DT, Wagner MJ, eds. Emergency and Primary Care of the Hand. Irving, TX, The American College of Emergency Physicians, 2001: 307-322.
44. Kaufman HD. The clinicopathological correlation of high-pressure injection injuries. Br J Surg 1968;55:214-218.
45. Harter BT Jr, Harter KC. High pressure injection injuries. Hand Clin 1986;2:547-552.
46. Hayes CW. High pressure injection injuries of the hand. South Med J 1982;75:1491-1498.
47. Schoo MJ, Scott FA, Boswick JA. High pressure injection injuries of the hand. J Trauma 1980;20:229-238.
48. Vasilevski D, Noorbergen M, Depierreux M, et al. High-pressure injection injuries to the hand. Am J Emerg Med 2000;18:820-824.
49. Lewis HG, Clarke P, Kneafsey B, et al. A 10-year review of high-pressure injection injuries to the hand. J Hand Surg 1998;23: 479-481.
50. Pinto MR, Turkula-Pinto LD, Cooney WP, et al. High-pressure injection injuries of the hand: A review of 25 patients managed by open wound technique. J Hand Surg 1993;18:125-130.
51. Kerr R. Diagnostic imaging of upper extremity trauma. Radiol Clin North Am 1989;27:891-908.
52. Jacobson JA, Powell, A, Craig JG, et al. Wooden foreign bodies in soft tissue: Detection at US. Radiology 1998;206:45-48.