Ventilator-Associated Methicillin-Resistant (MRSA) Pneumonia in a Patient with a Negative MRSA Nasal Swab

Background

Methicillin-resistant Staphylococcus aureus (MRSA) pneumonia can be community-acquired, healthcare-acquired, or associated with ventilator use. For ventilator-associated pneumonia (VAP), MRSA currently accounts for 20–40% of all cases [1]. There is no gold standard for diagnosing VAP; however, the most accurate diagnostics include histopathologic examination of lung tissue obtained by open lung biopsy, rapid cavitation of a pulmonary infiltrate in the absence of cancer or tuberculosis, positive pleural fluid culture, same species with same antibiogram isolated from blood and respiratory secretions without another identifiable source of bacteremia, and histopathologic examination of lung tissue at autopsy [2]. Given the invasive nature of lung biopsy and the infrequent occurrence of other manifestations used as diagnostics, less invasive testing has evolved to help in diagnosis and identification of a causative organism. Bronchoalveolar lavage (BAL), which typically requires bronchoscopy to obtain samples, is relatively invasive but has superior specificity over endotracheal aspirate culture (EAC) sampling [2,3]. Despite this, the improved specificity has not translated into improved outcomes for patients or less antibiotic use, and the Infectious Disease Society of America and American Thoracic Society currently recommend noninvasive techniques (eg, EAC) over invasive techniques (e.g. BAL) for diagnosis of routine VAP [3,4].

Even less invasive is the MRSA nasal swab. Originally produced to screen for MRSA carriers in the hospital setting, this test was not initially intended to guide antibiotic therapy for pneumonia [5]. However, it has become widely used in hospitalized patients in recent years to help “exclude” MRSA pneumonia due to the high negative-predictive value (NPV) of this test for pneumonia caused by MRSA bacteria [6–8]. A recent meta-analysis of studies on this topic found the MRSA nasal swab to have a pooled sensitivity and specificity of 70.9% and 90.3%, respectively, for MRSA pneumonia. With a reported 10% prevalence of potential MRSA pneumonia, the calculated positive-predictive value (PPV) was 44.8%, and the NPV was 96.5%. For the MRSA VAP subgroup, the pooled sensitivity, specificity, PPV, and NPV were 40.3%, 93.7%, 35.7%, and 94.8%, respectively [8].

Despite these impressive NPV numbers, false negatives will undoubtedly still occur. In clinical practice, this is less of an issue for medically stable patients on the medical wards. In these patients, antibiotics can simply be broadened if the patient fails to improve. In hemodynamically unstable patients in the Intensive Care Unit (ICU), however, failing to provide the correct antibiotic treatment could be fatal [6]. The present case report highlights the possibility of false negative results in this scenario and demonstrates that clinical judgement needs to be exercised when using MRSA nasal swab testing to de-escalate antibiotics, especially in the critically ill.

Case Report

A 58-year-old woman with a history of alcoholism and chronic obstructive pulmonary disease (COPD), with ongoing tobacco abuse, presented with shortness of breath and cough. Admission vital signs were significant for a respiratory rate of 32 breaths/min, temperature of 38.0°C, heart rate of 115 bpm, blood pressure of 73/40 mmHg, and oxygen saturation of 72% on room air. On physical exam, she was lethargic but arousable to voice. On auscultation of the heart, she was tachycardic but with regular rhythm and no murmurs. She was in mild respiratory distress with coarse rhonchi on auscultation of the lungs. Her extremities were warm with strong pulses. Her white blood cell count (WBC) was 12.4×103 cells/uL. Arterial blood gas demonstrated a pH of 7.07, pCO2 of 106 mmHg, p02 of 80 mmHg, and bicarbonate of 30 mmol/L. Chest X-ray (CXR) on admission did not demonstrate any acute pathology (Figure 1); however, computed tomography (CT) imaging of the chest revealed evidence of multifocal pneumonia (Figure 2). Our patient was diagnosed with septic shock secondary to community-acquired pneumonia (CAP) and started on ceftriaxone and azithromycin. She was fluid resuscitated with 3 liters of lactated ringers; however, her blood pressure continued to be low. She was then started on norepinephrine. Our patient’s SpO2 was still only 83% on 15 L of O2 via a non-rebreather mask, so she was intubated. While in the ICU, she initially improved after a 5-day course of ceftriaxone and azithromycin for treatment of CAP. Blood and respiratory cultures taken by endotracheal aspirate on admission were negative. She was liberated from norepinephrine on hospital day 2. Her fraction of inspired oxygen (Fi02) and positive end expiratory pressure (PEEP) were 40% and 5 cm H20 on hospital day 5 and 30% and 5 cm H2O on hospital day 6, respectively. She was successfully extubated in the morning on hospital day 6.

Later in the day on hospital day 6, she developed respiratory distress with a respiratory rate of 28 breaths/min, temperature of 38.2°C, heart rate of 112 bpm, blood pressure of 80/55 mmHg, and oxygen saturation of 85% on 60 L of heated high-flow oxygen therapy. Arterial blood gas demonstrated a pH of 7.32, pCO2 of 56 mmHg, pO2 of 78 mmHg, and bicarbonate of 24 mmol/L. She was re-intubated and restarted on norepinephrine. FiO2 and PEEP needs at this time were 100% and 5 cm H2O, respectively. A repeat CXR demonstrated a new left lower lobe infiltrate (Figure 3). VAP is defined as new pulmonary infiltrates on imaging plus clinical findings of infection. If extubation occurs, a new infection is considered to be ventilator-associated if re-intubation occurs within 1 calendar day. This was the case for our patient. She was initially started on vancomycin and cefepime for treatment of VAP. However, PCR testing of a MRSA nasal swab taken from the bilateral nares after re-intubation was negative, and vancomycin was discontinued on hospital day 7 when the results of this test were known. Our patient was then continued on cefepime monotherapy. Over the next 72 hours, she continued to experience fevers up to 38.2°C, an up-trending leukocytosis, and ongoing vasopressor need. At 72 hours after re-intubation (hospital day 9), a culture from an endotracheal tube aspirate, collected from the day of re-intubation, tested positive for 3+ MRSA bacteria. The organism was resistant to oxacillin, clindamycin, and erythromycin. It demonstrated sensitivity to linezolid, tetracycline, gentamycin, trimethoprim/sulfamethoxazole, and vancomycin. Blood cultures remained negative. Our patient was restarted on vancomycin. Over the course of the next 3 days, she improved and was liberated from both norepinephrine and the ventilator. She was transferred out of the ICU and completed a total of 7 days of intravenous vancomycin.

Discussion

The MRSA nasal swab has become widely used in recent years to help narrow antibiotic administration in hospitalized patients because of the test’s high NPV for pneumonia caused by MRSA bacteria [6–8]. Our case describes the limits of using the MRSA nasal swab for this purpose, which brings to attention several learning points for clinicians.

Our patient completed a course of antibiotics for CAP and subsequently developed VAP. Vancomycin was initially empirically started but was discontinued based on results of a negative MRSA nasal swab and relatively low local prevalence of invasive MRSA infections (incidence of 14.8 per 100,000 locally vs 20.1 per 100,000 nationally) [9,10]. Also notable, our patient did not have any other high-risk exposure history, such as recent hospital admissions or residence at an extended care facility. After 3 days of failing to improve on cefepime monotherapy, vancomycin was re-prescribed once respiratory cultures demonstrated 3+ growth of MRSA bacteria.

Negative results from MRSA nasal swabs have been shown to have high NPV for MRSA pneumonia in both CAP and VAP [6–8]. This test is appealing to providers as a clinical tool for a number of reasons. Firstly, vancomycin, which is commonly used to treat MRSA infections, can be significantly nephrotoxic [11].

Furthermore, MRSA as a causative organism of pneumonia can be difficult to exclude otherwise. The sensitivity of gram staining and culture of sputum and endotracheal aspirates is generally considered to be too low to be useful for NPV purposes [12,13]. Imaging can be a useful tool in CAP as MRSA pneumonia tends to cavitate and be bilateral [14–16]. However, in VAP, there are no radiographical features that reliably distinguish MRSA pneumonia from pneumonia caused by other organisms [16].

The MRSA nasal swab becomes less useful in areas with high local MRSA prevalence and in patients with a high pre-test probability of MRSA pneumonia, as it can lead to false negative results [7]. Also, when using the MRSA nasal swab to help narrow antibiotic administration in critically ill patients in the ICU, the chance of a false negative result can carry significant risks of further morbidity and mortality [6]. Thus, MRSA nasal swab testing should be combined with clinical judgement and used with caution in these populations.

To our knowledge, there are no published reports of similar cases. The only prospective study evaluating MRSA nasal screening for the purposes of de-escalating antibiotics was published in 2022 by Raush et al. They studied 523 critically ill patients and found no difference in hospital mortality, hospital duration, or rates of antibiotic re-initiation. However, only 2 of the 523 patients in this study met the clinical criteria for VAP, which limits the generalizability of this study to this patient population [17]. Current American Thoracic Society and Infectious Disease Society of America guidelines for CAP support withholding treatment for MRSA pneumonia if the nasal swab is negative, especially in non-severe CAP [18]. For VAP, the recommendations by these societies are less definitive, and they advise that use of the MRSA nasal swab for de-escalation of antibiotics should depend on local prevalence of MRSA respiratory infections [4].

Conclusions

This report highlights the need for caution when using the MRSA nasal swab to de-escalate antimicrobials. This test may be considered when deciding whether to de-escalate empiric MRSA antibiotics in populations with low MRSA prevalence. However, in critically ill patients with high risk and suspicion for MRSA pneumonia, clinicians should consider waiting until respiratory culture results are available before stopping empiric MRSA antimicrobial therapy.