Hypophosphatemia in meningococcal sepsis

Abstract

Sirs, We read with interest the article by Akil et al. [1], presenting a 3-year-old boy with fulminant meningococcemia and acute renal failure. In this patient hypophosphatemia occurred paradoxically in the presence of oliguria. We would like to further expand on the authors’ discussion about this electrolyte abnormality. The pathophysiological mechanisms of hypophosphatemia include internal redistribution (for example in cases of respiratory alkalosis or glucose administration), urinary loss, and decreased intestinal absorption [2]. Hypophosphatemia has long been reported for adult patients in the early phase of sepsis, most often caused by gram-negative bacteria [3]. Serum phosphate levels normalize within days, as the symptoms of acute septicemia subside. Acute renal loss of phosphate was thought to be an unlikely explanation, since many of these patients presented with oliguria or anuria. Severe hypophosphatemia was observed even in the presence of normal blood gases and prior to initiation of intravenous fluid administration [3]. Later, hypophosphatemia was also demonstrated for adult patients with pneumonia and other infections, mostly bacterial [4, 5, 6]. We reported that transient hypophosphatemia is a relatively frequent phenomenon in children admitted with bacterial infections [7]. In most of these studies, factors such as respiratory alkalosis, renal disease, prior intravenous fluid administration, and poor nutritional status were among the exclusion criteria or were not observed for many of the hypophosphatemic patients. The patient presented by Akil et al. [1] already had low serum phosphate (1.68 mg/dl) on admission, with normal blood urea nitrogen and creatinine levels (6 mg/dl and 0.27 mg/dl, respectively) [1]. So, what is the cause of hypophosphatemia in patients with acute infection? Barak et al. [6] investigated the relationship between serum phosphate and cytokine levels in adult patients with sepsis or other bacterial infections. They found that hypophosphatemia in these patients was associated with high levels of tumor necrosis factor-a (TNF-a) and interleukin-6 (IL-6). Injection of TNF-a or IL-6 in mice resulted in a marked decrease of their serum phosphate concentrations, suggesting a causative role for these cytokines in the pathogenesis of hypophosphatemia [6]. H glin et al. [5] reported that hypophosphatemic patients had a higher mean C-reactive protein level (CRP) than the normophosphatemic group. The prevalence of hypophosphatemia was higher in pediatric patients with elevated CRP levels, reaching 61% in children with pneumonia and CRP>150 mg/l [7]. Hypophosphatemia occurred before the increase in CRP and resolved before the normalization of CRP levels [7]. These findings were in agreement with those of Barak et al. [6], since TNF-a and IL-6 are the main stimuli for the production of CRP from hepatocytes, and their peak levels in plasma precede those of CRP by 24–48 h. Consequently, the hypophosphatemia in the presented case of meningococcal sepsis could be attributed to the overproduction of cytokines such as TNF-a and IL-6, which are known to play a major role in the pathogenesis of sepsis. Notably, the patient’s CRP level on admission was very high (245 mg/l) [1]. In parallel with these reports, other studies have investigated urinary electrolyte excretion in patients with sepsis. Kalan et al. [8] reported increased urinary loss of phosphate in septic neonates. An increased parathyroid hormone secretion was found in critically ill adult patients, some with septicemia, but was not associated with hypophosphatemia or changes in urinary calcium excretion [9]. There is also a possibility of inadequate or unbalanced (in relation to calcium) phosphate supplementation as a cause of hypophosphatemia associated with infection, C. Antachopoulos First Department of Pediatrics, Athens University, Aghia Sophia Children’s Hospital, Athens, Greece