Infectious Complications of Drug Abuse

Afroza Liton

William L. Marshall

Drug abuse, the deliberate taking of an unprescribed drug dose or illicit substance, is a pervasive problem in our society [1]. A variety of drugs are abused, including opiates, depressants, stimulants, and hallucinogens. This chapter will focus on infections that occur as a consequence of drugs that are either explicitly illegal or those which are legal but are used by the patient for purposes other than for which they were prescribed. Abused drugs can be administered by a variety of means, including “snorting” through the nasal mucosa, via inhalation through smoking, and orally by parenteral routes, including injection into the soft tissues, called “skin popping,” or directly into the vascular system.

Drug abuse is attended by an increased risk in a number of infections, some of which may lead patients to be admitted to the intensive care unit (ICU) [2]. Infections associated with parenteral drug abuse include skin and soft tissue infection, endocarditis, bone and joint infections, pneumonia, ophthalmologic infections, and hepatitis [2,3,4]. Illicit drugs are often “cut” or mixed with adulterants, which may be contaminated with bacteria or may suppress the immune response—as is the case with agranulocytosis caused by levamisole-containing cocaine leading to bacterial or fungal infection [5]. Illicit drug injection occurs under unsanitary conditions, using drugs that are not sterile and injection equipment that has often been used more than once. Such practices provide a mechanism for passage of a variety of infectious agents. Although in some instances, particularly for the hepatitis viruses and human immunodeficiency virus (HIV), the infectious agent is passed directly from blood-contaminated drug paraphernalia to the patient, the mode of spread is less clear for other agents. Prevention of infectious complications of drug use is directed at treating addiction, or failing that, mitigating infectious complications via needle exchange programs [6]. Finally, many patients with substance abuse problems are homeless, have poor nutrition, and live under crowded conditions, placing them at increased risk for tuberculosis.

Fever

Fever is one of the most common complaints of parenteral drug users presenting to the hospital. Self-limited illnesses are the most common causes of fever in this population. More significant etiologies include pneumonia, cellulitis, and soft tissue abscesses. Endocarditis accounts for fewer than 15% of all cases of fever [7].

All febrile parenteral drug users should undergo a thorough history, physical examination, and have routine blood laboratories and chest radiographs taken. Particular attention should be paid to abnormalities of the skin and soft tissues, cardiac valvular abnormalities, bony tenderness, and pulmonary abnormalities. However, clinical evaluation alone often does not differentiate major disease from trivial illness in these patients. Parenteral drug users who are febrile should be admitted to the hospital for further observation.

Weisse et al. have developed an algorithm for febrile parenteral drug abusers with no apparent source of infection [7]. In this approach, blood cultures are obtained on all patients and empiric antibiotic therapy is started. If blood cultures are positive or if the patient has clinical stigmata indicative of endocarditis, an echocardiogram is performed. If valvular vegetations are seen, the diagnosis of endocarditis is considered established. On the other hand, if blood cultures are negative and the patient is clinically well, antibiotic therapy may be stopped. However, parenteral drug users commonly self-administer antibiotics and this practice may substantially reduce the likelihood of positive blood cultures, as can prophylactic antibiotics in HIV+ patients [8,9]. Hence, careful clinical evaluation is advised when making antibiotic decisions in these patients.

Bacteremia

Bacteremia is a frequent occurrence in the febrile parenteral drug user [10,11]. Approximately 60% of bacteremias in parenteral drug abusers are due to causes other than endocarditis [12]. Of these, the majority are due to either skin or soft tissue infections or to mycotic aneurysms of peripheral arteries. A smaller number of bacteremias are due to miscellaneous causes, such as septic arthritis, septic thrombophlebitis, or pneumonia. In about 3% of cases, the source of the bacteremia is undiscovered.

Although the organisms associated with bacteremias in the parenteral drug user may vary based on geographic location and the type of drug abused, some generalizations can be made [12,13]. Drug users have an increased incidence of staphylococcal carriage of the skin, nose, and throat [14]. Bacterial infection derives principally from the user’s own flora, so that Staphylococcus aureus constitutes the majority of bacteremias in these patients. In this regard, methicillin-resistant S. aureus (MRSA) infections are now being encountered with increasing frequency in parenteral drug users and in the community [14,15].

Streptococci and Gram-negative aerobic bacilli are the next most frequently isolated organisms. Polymicrobial bacteremias occur in about 10% of cases, and in about two-thirds of these cases at least one of the organisms isolated is a Staphylococcus spp [12]. Bacteremia and other infections due to the facultative anaerobe Eikenella corrodens are particularly associated with injecting drug users who contaminate the injection needle or the injection site with saliva [16].

The approach toward the bacteremic parenteral drug user should be to search for an underlying etiology and to begin empiric antibiotic treatment. The isolation of a group A β-hemolytic streptococci from the blood should prompt a search for a cutaneous or soft tissue focus of infection [17]. Empiric antibiotic therapy may be based on local experience but should generally include agents directed against staphylococci and streptococci as well as aerobic Gram-negative bacilli. If
MRSA infections have previously occurred in parenteral drug users in the community, vancomycin should be considered.

Soft Tissue Infections

Skin and soft tissue infections occur commonly in the parenteral drug user and are increasing in frequency [18,19]. Such infections are often polymicrobial and appear to derive from either the skin or oral cavity [18,20,21]. The most common pathogens are S. aureus, streptococci, oral anaerobes, and aerobic Gram-negative bacilli [16,17,18,19,20,21]. Cutaneous infection in the intravenous drug user generally occurs in the antecubital fossa, forearm, and hand since these are the sites of the most accessible veins. However, intravenous drug users may also avail themselves of other, less available sites with infection occurring in the feet, legs, anterior neck, groin, and axilla [22,23].

The most common skin infections in the injecting drug user (IDU) are simple cellulitis and localized skin abscess. These occur more frequently among those who “skin pop” compared to those who inject intravenously [24]. Simple cellulitis usually requires only antibiotic therapy directed against staphylococci and streptococci. Since the incidence of MRSA infections is rising, and the IDU is particularly at risk for MRSA infections of the skin and soft tissues [14], patients requiring intravenous therapy should receive vancomycin. Localized soft tissue abscesses that do not penetrate into the deep subcutaneous tissue should be drained. Given the risk of occult bacteremia in this population, antibiotic therapy should be given as directed by Gram stain of the drained material. In all patients with a history of injection drug use, blood cultures should be obtained in the workup of skin and soft tissue infections.

The presence of vesicles or bullae, an area of central necrosis within a larger area of erythema, and the presence of subcutaneous crepitation in a patient with systemic toxicity is suggestive of necrotizing fasciitis [25]. Gas seen in the soft tissues on radiographs is also indicative of deep infection [26]. However, extensive necrosis may be present even in the absence of these signs, and surgical exploration should be considered in any case that manifests local erythema, fluctuance, and induration [27]. Suspicion for needles or other foreign bodies should similarly prompt surgical exploration. Any abnormal material from this exploration should be immediately examined using Gram stain to provide the basis for empiric antimicrobial therapy. Examination of a sample of tissue using frozen-section biopsy may also be useful [28]. Magnetic resonance imaging (MRI) typically reveals increased T2 signal along fascial planes and gadolinium enhancement, whereas contrast-enhanced computed tomography (CT) scanning is a less sensitive diagnostic tool for necrotizing fasciitis [26].

Necrotizing fasciitis, pyomyositis, or gangrene requires immediate, aggressive debridement in the operating room in association with parenteral antibiotics [27]. Gram stain and culture are imperative to guide antimicrobial therapy. Empiric therapy should be directed against staphylococci, streptococci, anaerobes, and aerobic Gram-negative bacilli. Surgical debridement may be required on multiple occasions before infection is controlled [29]. There have been multiple outbreaks of soft tissue infection with or without systemic symptoms associated with Clostridium spp discussed later in this chapter.

Peripheral Vascular Infections

Because parenteral drug use often involves vascular injection of material under non-sterile conditions, it is not surprising that a wide range of vascular complications may result from these practices [30]. The most frequent manifestations of such infections are fever associated with pain, redness, and swelling over the involved area. When the injecting site is into the deep tissues of the groin or neck, it may be difficult to distinguish involvement of vascular structures from simple cellulitis, soft tissue abscess, or fasciitis. If there is any question, angiography should be performed to determine if vascular tissue is involved. Septic thrombophlebitis usually presents as fever, bacteremia, and swelling over the involved vein. This can often be treated with antibiotics alone, although incision, drainage, and removal of the vein are sometimes necessary. Anticoagulation is generally not required [12].

Mycotic aneurysms result when the user injects directly into the artery [12,30]. Aneurysms most frequently occur in the femoral arteries. Carotid aneurysms and brachial artery aneurysms occasionally occur [30]. The classic presentation of this syndrome is a febrile patient with a tender, pulsatile mass, usually in the groin or the neck. Sometimes, there is a small amount of bleeding at the site. If there is any question of an aneurysm, a vascular surgical consultation should be obtained prior to any exploration of the lesion. Angiography will confirm the site and extent of the aneurysm. The most frequent microbiological agents isolated are S. aureus and streptococci, with aerobic Gram-negative bacilli occasionally being identified [12]. Empiric antibiotic therapy should be directed against these organisms. Ligation and excision of the involved arterial segment is usually successful [31].

Endocarditis

Endocarditis in the parenteral drug abuser differs in several respects from endocarditis in the nonaddict. It is more likely to occur in persons without underlying valvular heart disease, to involve the tricuspid valve, to be due to S. aureus, and to have a more benign outcome [3,32]. Certain types of intravenous drug abuse may predispose to the development of endocarditis. Heroin use has long been associated with this complication [33].

Tricuspid-valve endocarditis is the prototypical presentation of endocarditis in the parenteral drug user [33]. The patient complains of fever, usually for less than 1 week. There may be a history of chills and pleuritic chest pain and occasionally hemoptysis. On physical examination, fever is a nearly universal finding. A systolic murmur may or may not be present on admission, but often develops during the course of therapy. Signs of peripheral embolization, such as petechiae, splinter hemorrhages, Janeway lesions, or Roth spots, are uncommon. Osler’s nodes are frequently absent. On chest radiograph, multiple patchy infiltrates indicative of pulmonary emboli are strongly suggestive of the diagnosis of tricuspid endocarditis. Blood cultures are usually positive and in the majority of instances, S. aureus is isolated. When blood cultures are negative in the face of the appropriate clinical syndrome, one should suspect that the patient has recently taken antibiotics.

Endocarditis involving the valves of the left side of the heart may also occur in the parenteral drug user. Compared to patients with tricuspid-valve endocarditis alone, there is more likely to be a history of underlying heart disease [12]. On examination, a heart murmur is usually evident on presentation, and peripheral emboli are frequent. Streptococci are more likely to be isolated from the blood, but S. aureus is still frequently isolated [3,12,33].

In addition to staphylococci and streptococci, a variety of other organisms have been associated with endocarditis in the parenteral drug user, including aerobic Gram-negative bacilli, particularly P. aeruginosa, and fungi, notably Candida spp [3]. Moreover, polymicrobial bacteremia is a well-recognized complication of endocarditis in this population and is usually
indistinguishable on clinical grounds from that due to a single organism [34].

Only gold members can continue reading. Log In or Register to continue