University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Management Recommendations > Bacterial Infections

Ward 86 Management Recommendations

Prevention and Treatment of Serious Bacterial Infections in Patients with HIV Disease: Community-Acquired Pneumonia and Skin/Soft Tissue Infections*

updated August 2018

Contributors: Mark A. Jacobson, MD
Lisa Winston, MD
Annie Luetkemeyer, MD
Monica Gandhi, MD

Community-Acquired Pneumonia (CAP)

  1. HIV-infected patients, even those with high CD4 counts and suppressed on antiretroviral therapy (ART), are at high risk of developing CAP. In a large, population-based, cohort study, hospitalization for pneumonia occurred 5 times more often in HIV-infected patients on ART with CD4 counts of >300 cells/µL than in controls who were not HIV infected and 20 times more often in AIDS patients with a CD4 count of <100 cells/µL.(1)

  2. In HIV-infected patients, the majority of CAP cases are caused by invasive Streptococcus pneumoniae infection. Pneumococcal CAP may not appear as focal infiltrates in patients with advanced AIDS but instead as diffuse interstitial infiltrates, mimicking Pneumocystis pneumonia. Outcomes of standard antibiotic therapy for community-acquired pneumococcal pneumonia are no worse in HIV-infected than HIV-uninfected patients.

  3. Outpatient therapy. HIV-infected patients who meet standard criteria for outpatient antibiotic therapy can be treated empirically as outpatients with the following caveats:

    1. Daily levofloxacin or moxifloxacin is first-line therapy but should be avoided if pulmonary tuberculosis is being considered in the differential diagnosis. Quinolone-alone treatment of active tuberculosis may result in initial clinical improvement of tuberculosis, which could delay the diagnosis.

    2. Amoxicillin is a reasonable alternative. Since amoxicillin half-life is long enough for BID dosing, we give 1 gm BID to facilitate adherence. Doxycycline or a macrolide such as azithromycin can be added to cover atypical organisms (Legionella, Mycoplasma, or Chlamydia species), although there is some data suggesting that atypical organisms are uncommon causes of CAP in HIV-infected patients.

    3. We have seen a number of treatment failures with the use of doxycycline or azithromycin alone for CAP. This may be because 20-30% of S pneumoniae species obtained from hospitalized patients in North America and Europe are resistant to these drugs.

    4. We recommend that antibiotics for outpatient CAP be given for a 7-day course because high relapse rates have been reported to occur in AIDS patients with invasive pneumococcal infection.

  4. Inpatient therapy. We initiate IV ceftriaxone or levofloxacin for HIV patients who are sick enough to require hospitalization and then switch to oral therapy, as above, once they have clinically improved. We recommend a 7-day antibiotic course for patients with uncomplicated pneumonia, 10-14 days for patients with multilobar pneumonia or severe hypoxia. Of note, two recently published randomized trials reported improved outcome in patients hospitalized with CAP who were given adjunctive corticosteroid therapy. However, the generalizability of this to HIV-infected patients is still unknown and awaits further data from ongoing trials.

  5. Prevention. We have been unimpressed by the 23-valent pneumococcal vaccine (Pneumovax) clinical efficacy data from the studies that are generalizable to adult HIV-infected patients. However, a new 13-valent pneumococcal vaccine, Prevnar 13, with improved immunogenicity, has been approved for use by adults in the United States. Some of us favor a prime-boost vaccination strategy of treating patients with ART until their plasma HIV RNA is undetectable, then giving a dose of Prevnar 13 followed 2-6 months later with boosting dose of Pneumovax. Others favor priming with the Prevnar vaccine as early as possible once a patient enters care, then boosting with Pneumovax after the patient has completed 2-6 months of ART. The rationale for this latter strategy is that, even though primary vaccination may be less effective when a patient's T-cell counts are low and HIV replication is not suppressed, those with low T-cell counts and uncontrolled HIV replication also are at the highest risk of developing invasive pneumococcal disease. The CDC recommends that those who have previously received Pneumovax should not be given a Prevnar dose until ≥1 year has passed after the last Pneumovax dose was administered.

Skin and Soft Tissue Infection (SSTI)

  1. A majority of SSTIs in HIV patients are caused by Staphylococcus aureus or Group A Streptococcus (GAS). Careful clinical examination may suggest which organism is the more likely cause, a distinction that can be helpful in choosing antibiotic therapy. Signs suggesting S aureus SSTI are abscess or furuncle formation, folliculitis, or cellulitis with a central, hair follicle-based lesion. GAS infection can cause cellulitis but rarely leads to abscess or furuncle formation, though it can progress to necrotizing fasciitis. GAS cellulitis often occurs in the extremities and may be accompanied by lymphangitic streaks progressing proximally.

  2. Methicillin-resistant S aureus (MRSA). In many areas of the United States, a large proportion of S aureus SSTIs are caused by MRSA. Most of these are local infections (abscesses, furuncles, or folliculitis). The frequency at which MRSA and MSSA cause nonpurulent cellulitis is unclear, as blood cultures are usually negative and cultures from skin are not helpful. However, adding trimethoprim-sulfamethoxazole (TMP-SMX) to a first-generation cephalosporin to cover MRSA infection is not associated with a statistically significant benefit in the general population with nonpurulent cellulitis.(2,3) Of note, men who have sex with men (MSM) have an increased incidence of serious MRSA infection caused by USA300 strain.(4)

  3. Abscesses and furuncles require treatment with incision and drainage (I&D). Adjunctive antibiotic therapy targeting MRSA also should be strongly considered, even in the absence of surrounding cellulitis. For treatment of uncomplicated abscesses without surrounding cellulitis, studies of I&D alone have demonstrated high cure rates.(5) However, several recent multicenter, randomized, placebo-controlled trials have reported:

    1. Higher cure rates for abscesses ≤5 cm in diameter when adjunctive TMP-SMX 160/800 mg PO BID or clindamycin 300 mg PO TID for 10 days was initiated after I&D.(6)
    2. Lower rates of subsequent hospitalization, need for surgery, and new infections when TMP-SMX 160/800 mg PO BID for 1 week was initiated after I&D.(7)

    Although there are no data from randomized trials regarding doxycycline, this antibiotic at a 100 mg BID dosage also may be effective for patients who cannot tolerate TMP-SMX or clindamycin.

  4. Patients with clinical findings of SSTI who have fever or other signs of sepsis should have blood cultures obtained and be hospitalized for empiric IV antibiotic therapy. In areas where MRSA is prevalent, empiric therapy should include vancomycin or some other agent with bactericidal activity against MRSA.

  5. Cellulitis without signs of sepsis may be treated on an outpatient basis with oral bacteriostatic antibiotics, rest, and elevation of the affected area.

    1. Oral antibiotics that are active against both methicillin-sensitive S aureus (MSSA) and MRSA include TMP-SMX, clindamycin, doxycycline, and linezolid. Because of linezolid's high cost and side-effect profile, we use this agent only when resistance or intolerance to all other agents has been documented.

    2. Oral antibiotics with activity against GAS include amoxicillin, first-generation cephalosporins (eg, cephalexin), and clindamycin.

    3. We treat cellulitis until there is clinical resolution of signs, at a minimum for 5 days.

    4. An outpatient antibiotic regimen for cellulitis should cover both S aureus and GAS unless there is an obvious central abscess or furuncle indicating S aureus is the cause, in which case treatment for GAS can be foregone. Considerations in choosing outpatient antibiotic regimens are as follows:

      1. The typical TMP-SMX dosage is 1 double-strength tablet BID. TMP-SMX will cover most MSSA and MRSA strains but may be insufficient for GAS, though in a recent multicenter trial that randomized patients with uncomplicated SSTI to clindamycin versus TMP-SMX, there was no significant difference in cure rate among the 110 evaluable subjects who had cellulitis in the absence of abscess.(8)

      2. Doxycycline 100 mg PO BID will cover most MSSA and MRSA strains but is insufficient for GAS.

      3. A first-generation cephalosporin (eg, cephalexin) 500 mg QID will cover MSSA and GAS but not MRSA.

      4. Clindamycin 300 mg BID** will cover most MSSA and MRSA strains. However, one third to one half of the GAS isolates tested in our hospital laboratory over the last 2 years were resistant to clindamycin, a substantially higher proportion than we have seen previously. Clindamycin-resistant GAS is also emerging in several European countries. Hence, we now prefer to use another drug for GAS coverage in outpatient cellulitis treatment (eg, amoxicillin or a first-generation cephalosporin) whenever there is not an obvious central abscess or furuncle indicating S aureus to be the only cause.

      5. Amoxicillin 875 or 1,000 mg PO BID will cover GAS but not MSSA or MRSA.

      6. Maximum coverage in situations wherein the causative pathogen is not obvious is a combination of two antibiotics: either amoxicillin 875 or 1,000 mg BID or a first-generation cephalosporin 500 mg QID PLUS either TMP-SMX 1 double-strength tablet BID or doxycycline 100 mg BID. (We avoid combining clindamycin with amoxicillin or a cephalosporin because of the increased risk of antibiotic-associated diarrhea.)

  6. Patients with extensive folliculitis may benefit from low-dose oral antibiotic therapy targeting MRSA and administered until there is a sustained response, usually for 2-8 weeks. We have had success with clindamycin 150 mg daily**, doxycycline 100 mg BID, and TMP-SMX 1 double-strength tablet BID regimens.

  7. Patients with frequently recurrent abscesses may benefit from prophylactic low-dose oral antibiotic therapy targeting MRSA as described above for folliculitis and administered for a period of 3 months.

  8. Another approach to patients with frequently recurring folliculitis or abscesses is to attempt to decolonize the skin. Although randomized trials have not shown benefit of such a strategy, some experts have recommended a 5-day trial of twice-daily intranasal mupirocin, daily chlorhexidine washes, and daily decontamination of personal items such as towels, sheets, and clothes. We have had some success with daily dilute bleach baths in which the patient immerses in a full tub of water to which 1/4 to 1/2 cup of bleach has been added.

* Management of diarrhea caused by bacterial pathogens is covered in the section Diagnosis and Management of Diarrhea.

** For suspected MRSA SSTIs appropriate for outpatient antibiotic treatment, we dose clindamycin less frequently than recommended in many guidelines because: 1) oral administration of clindamycin achieves high drug concentrations in infected skin and subcutaneous tissue, 2) clindamycin has a long duration of antibactericidal activity in serum against S aureus with BID dosing, 3) patients with advanced HIV disease given oral clindamycin actually may have higher serum drug levels than otherwise healthy individuals, 4) HIV patients have an increased risk of C difficile-associated disease, for which clindamycin exposure is a risk factor, and 5) in our experience, HIV patients often have more non-C difficile-associated diarrhea with TID and QID dosing than with BID dosing. Anecdotally, we have observed comparable efficacy with TMP-SMX, doxycycline, and clindamycin for outpatient MRSA-associated SSTIs. However, an ongoing NIH-sponsored, multicenter, randomized trial is testing the optimal antibiotic choice for the outpatient treatment of MRSA SSTIs and should provide better guidance in the near future.


  1. Sogaard OS, Lohse N, Gerstoft J, et al. Hospitalization for pneumonia among individuals with and without HIV infection, 1995-2007: a Danish population-based, nationwide cohort study. Clin Infect Dis. 2008 Nov 15;47(10):1345-53.
  2. Moran GJ, Krishnadasan A, Mower WR, et al. Effect of cephalexin plus trimethoprim-sulfamethoxazole vs cephalexin alone on clinical cure of uncomplicated cellulitis: a randomized clinical trial. JAMA. 2017 May 23;317(20):2088-96.
  3. Pallin DJ, Binder WD, Allen MB, et al. Clinical trial: comparative effectiveness of cephalexin plus trimethoprim-sulfamethoxazole versus cephalexin alone for treatment of uncomplicated cellulitis: a randomized controlled trial. Clin Infect Dis. 2013 Jun;56:1754-62.
  4. Diep BA, Chambers HF, Graber CJ, et al. Emergence of multidrug-resistant, community-associated, methicillin-resistant Staphylococcus aureus clone USA300 in men who have sex with men. Ann Intern Med. 2008 Feb 19;148(4):249-57.
  5. Rajendran PM, Young D, Maurer T, et al. Randomized, double-blind, placebo-controlled trial of cephalexin for treatment of uncomplicated skin abscesses in a population at risk for community-acquired methicillin-resistant Staphylococcus aureus infection. Antimicrob Agents Chemother. 2007 Nov;51(11):4044-8.
  6. Daum RS, Miller LG, Immergluck L, et al. A placebo-controlled trial of antibiotics for smaller skin abscesses. N Engl J Med. 2017 Jun 29;376(26):2545-55.
  7. Talan DA, Mower WR, Krishnadasan A, et al. Trimethoprim-sulfamethoxazole versus placebo for uncomplicated skin abscess. N Engl J Med. 2016 Mar 3;374(9):823-32.
  8. Miller LG, Daum RS, Creech CB, et al. Clindamycin versus trimethoprim-sulfamethoxazole for uncomplicated skin infections. N Engl J Med. 2015 Mar 19;372(12):1093-103.