Duration of viral shedding in hospitalized patients infected with pandemic H1N1

Abstract

Background: The first influenza pandemic of the 21th century was ignited by a new strain of influenza A virus (A/H1N1pdm). Specific patient groups, including those with comorbidities, pregnant women, young children, older and immunocompromised patients, are at increased risk for serious influenza-related disease. This study was aimed at investigating the influence of clinical presentation, antiviral treatment and possible drug resistance-associated mutations, on the extent and duration of viral shedding in patients infected with A/H1N1pdm.

Methods: An observational study was performed, based on retrospective review of clinical and laboratory records of patients who were hospitalized for A/H1N1pdm infection at the National Institute for Infectious Diseases "L. Spallanzani", Rome, Italy, between April 24 and December 31, 2009. Among 119 hospitalized patients, 39 were selected for a post hoc analysis, based on the availability of serial nasopharyngeal swabs samples and related information.

Results: Eleven out of the 39 study patients (28.2%) presented with pneumonia; 29 (74.4%) received antiviral treatment. Patients with pneumonia were significantly older than patients without pneumonia. The mean values of viral RNA concentration were not significantly increased in patients with pneumonia, but a significant increase in the duration of viral shedding was observed as compared to patients without pneumonia. In patients receiving antivirals, the viral RNA concentration was significantly reduced in comparison to untreated patients at days 4-5 after symptom onset, while the overall duration of viral shedding was only marginally affected. A significant correlation between duration of viral shedding and time elapsed between symptom onset and therapy start was observed, with a significant reduction of days of viral shedding when therapy was initiated within 2 days of symptoms appearance. No known drug resistance mutations were detected in patients with prolonged viral shedding.

Conclusions: Our results show that severe respiratory illness is associated with delayed virus clearance in patients with A/H1N1pdm infection. Antivirals caused an early reduction of viral load, but only marginally affected the overall duration of shedding. Prolonged shedding was not associated with the emergence of strains carrying known drug-resistance mutations.

Figure 1

Correlation of initial viral load with the duration after symptom onset in 533 patients. The analysis was performed on 533 NPS collected during the first visit and sent to the Virology Laboratory of the National Institute for Infectious Diseases "L. Spallanzani", Rome, Italy for laboratory diagnosis (r = -0.203, p < 0.0001).

Figure 2

Time course of viral load and duration of viral shedding according to presence of pneumonia and antiviral treatment. Panels A and C: mean values and standard deviation (represented by bars) of influenza A/H1N1pdm in NPS at various time points after symptom onset; panels B and D: proportion of samples positive to influenza A/H1N1pdm PCR according to time from symptom onset. Panels A and B: patients with pneumonia vs. without pneumonia; Panels C and D: treated vs. untreated patients. In panel A the number of samples collected at days 0-1, 2-3, 4-5, 6-7, 8-9 and ≥10 were (for patients with pneumonia) 3, 8, 8, 9, 5, and 21 and (for patients without pneumonia) 8, 21, 30, 7, 14 and 10. In panel C the number of samples collected at days 0-1, 2-3, 4-5, 6-7, 8-9 and ≥10 were (for treated patients) 9, 24, 24, 23, 13 and 29, and (for untreated patients) 2, 5, 14, 13, 6 and 2. * p ≤ 0.010, ** p ≤ 0.010, *** p ≤ 0.010; §: p = 0.002.

Figure 3

Correlation of duration of viral shedding with interval between symptom onset and therapy start. The analysis was performed on data from 20 patients treated with inhibitors of NA, who reached undetectable PCR during the observation period (r = 0.531, p = 0.016).