Fever Resolution in Pyelonephritis: Be Patient
Abstract & Commentary
Synopsis: Fever persistence beyond 96 hours suggests the need for infectigation of possible complications.
Source: Behr MA, et al. Fever duration in hospitalized acute pyelonephritis patients. Am J Med 1996;101:277-280
Behr and colleagues report the results of a retrospective chart review of 234 patients hospitalized for a suspected urinary tract infection (UTI) at St. Mary’s Hospital in Montreal. Of those 234, 70 patients met a definition of acute pyelonephritis, defined as the presence of fever greater than 38.0°C oral and a urine containing greater than 105 colony forming units. There were nearly three times as many women as men. Thirty of 70 (42.95) had previous UTI. Only 7% were associated with stones, and only 4.3% of the patients were pregnant. Escherichia coli was the causative organism in 805 of the cases. The mean duration of fever was 39 hours and a median was 34 hours. After 48 hours of hospitalization, more than one-fourth were still febrile, and by 72 hours, 13% still had persistent fever. The 95th percentile for fever duration was 92 hours. The longest duration of fever (225 hours) was in a 25-year-old catheterized man growing Enterococcus faecalis.
Radiologic studies were performed in 43 patients ; 7 of 43 (16%) had an obstructive urinary tract abnormality, although none of the patients had intrarenal or perinephric abscesses. This frequency of complicated cases is similar to that reported in three other published series of radiologic findings in pyelonephritis. Of note is for that the median duration of fever was not prolonged for the seven patients with obstruction, In all 70 patients, the disease resolved without clinically evident complications.
When multivariate analysis was applied to the data, five variables were associated with duration of fever: isolation of E. coli, female gender, leukocytosis, decreased age, and increased creatinine.
COMMENT BY JOSEPH JOHN, MD, and ALEXANDER ACKLEY, MD
This is the largest published series describing the natural history of treated pyelonephritis. Infectious disease physicians tend to see pyelonephritis patients who have an inordinate lengthy duration of fever. Behr and colleagues document that fever can often persist up to four days. Indeed, they found that more than 10% of patients still had fever at 72 hours, although all but one (the catheterized male with enterococcal infection) resolved fever by 108 hours.
Managed care companies sifting through these data could rightfully force our hand further with rapid release from the hospital of patients with pyelonephritis. At the same time, the data help us to formulate a concrete recommendation to discharged patients: If the patient, once home, has fever more than four days after admission, further workup is warranted, including an ultrasound of the kidneys.
The authors, unfortunately, were not able to analyze the role that different antibiotics may play in resolution of fever since there were very few resistant strains of bacteria isolated. The frequency of resistant enteric bacilli continues to increase on the North American Continent. It would be interesting to repeat such a study in a few years when the impact of resistant strains can be factored into the analysis of prolonged fever in pyelonephritis. (Dr. Ackley is Professor of Clinical Medicine, Robert Wood Johnson Medical Medical Center, Princeton, NJ.)