Improving care and outcomes of inpatients with syndrome of inappropriate antidiuresis (SIAD): a prospective intervention study of intensive endocrine input vs. routine care

Endocrine DOI 10.1007/s12020-016-1161-9 ORIGINAL ARTICLE Improving care and outcomes of inpatients with syndrome of inappropriate antidiuresis (SIAD): a prospective intervention study of intensive endocrine input vs. routine care Ploutarchos Tzoulis1 Helen Carr1 Emmanouil Bagkeris2 Pierre Marc Bouloux1 ● ● ● Received: 29 August 2016 / Accepted: 25 October 2016 © The Author(s) 2016; This article is published with open access at Springerlink.com Abstract Purpose: The syndrome of inappropriate antidiuresis is often undertreated with most patients discharged with persistent hyponatraemia. This study tested the hypothesis that an endocrine input is superior to routine care in correcting hyponatraemia and can improve patient outcomes. Methods: This single-centre prospective-controlled intervention study included inpatients admitted at a UK teaching hospital, with serum sodium ≤ 127 mmol/l, due to syndrome of inappropriate antidiuresis over a 6-month period. The prospective intervention group (18 subjects with mean serum sodium 120.7 mmol/l) received prompt endocrine input, while the historical control group (23 patients with mean serum sodium 124.1 mmol/l) received routine care. The time needed for serum sodium increase ≥ 5 mmol/l was the primary endpoint. Results: The intervention group achieved serum sodium rise by ≥5 mmol/l in 3.5 vs. 7.1 days in the control group (P = 0.005). In the intervention group, the mean total serum sodium increase was 12 mmol/l with only 5.8 % of patients discharged with serum sodium < 130 vs. 6.3 mmol/l increase (P < 0.001) and 42.1 % of the subjects discharged with serum sodium < 130 mmol/l in the control group (P = 0.012). The mean length of hospital stay in the intervention group (10.9 days) was significantly shorter than in the control group (14.5 days; P = 0.004).The inpatient mortality * Ploutarchos Tzoulis 1 Centre for Neuroendocrinology, Royal Free Campus, University College London Medical School, London NW3 2QG, UK 2 Centre of Epidemiology and Biostatistics, Institute of Child Health, University College London, London, UK rate was 5.5 % in intervention arm vs. 17.4 % in control arm, but this difference was not statistically significant. Conclusions: Since the endocrine input improved time for correction of hyponatraemia and shortened length of hospitalisation, widespread provision of endocrine input should be considered. Keywords Hyponatraemia Syndrome of inappropriate antidiuretic hormone secretion Vasopressin SIADH Sodium ● ● ● ● Introduction Hyponatraemia is associated with considerable morbidity and mortality [1–4], and carries a substantial economic burden related to increased hospital resource utilisation, length of hospital stay and risk of readmission [5, 6]. Syndrome of inappropriate antidiuresis (SIAD), the commonest cause of hyponatraemia [7], is often undertreated with recent large observational studies showing that most patients are discharged with persistent hyponatraemia [8–10]. Despite suboptimal standards of care for SIAD, no prospective studies have examined the effect of widespread provision of endocrine input on its management. Also there is paucity of data about whether correcting hyponatraemia can improve patient-related outcomes, such as mortality, length of hospital stay, and readmission rate. The primary hypothesis of our study was that prompt and intensive endocrine input was superior to non-specialised ‘standard’ clinical care in correcting hyponatraemia with the primary endpoint being time to achieve an increase in serum sodium (sNa) by ≥5 mmol/l. Secondary objectives were: Endocrine (1) to compare the total rise in sNa and the percentage of patients discharged with sNa < 130 mmol/l between intervention and control group (2) to examine the effect of endocrine input on inpatient mortality and length of hospital stay (3) to study whether correction of hyponatraemia improves cognitive function. Materials and methods serum creatinine >200 umol/l or receiving renal replacement therapy, (7) uncontrolled hyperglycaemia with serum glucose >15 mmol/l, (8) pregnancy/breastfeeding, (9) receiving end-of-life care. Taking into account preliminary data from our cohort indicating a mean time of 5.5 days to reach the primary endpoint of sNa rise ≥5 mmol/l and if standard deviation (SD) for each arm is 1, power sample size was estimated, as 18 patients in each arm, in order to show 20 % difference in the primary endpoint (1.1 day) with 90 % power and 5 % significance level. Study design Control group This single-centre prospective-controlled intervention study was conducted in a UK teaching hospital over a 6-month period. During the first 3-month period (1st October 2014–31st December 2014), all patients underwent ‘routine’ care (control group), while in the following 3-month period (1st January 2015–31st March 2015), patients received intensive endocrine input (intervention group). The rationale behind the use of a historical control was that undertaking a randomised controlled trial in a real-life clinical setting was considered impractical since clinicians might often seek expert help from the investigators about patients allocated to the control arm, resulting in possible ‘dilution’ of the control group with patients receiving the intervention. The time required to achieve sNa increase of ≥5 mmol/l was chosen as the primary endpoint since this magnitude of correction can be sufficient to improve symptoms of hyponatraemia [11]. The study received ethical approval from the London-Camden & Islington Research Ethics Committee, and all subjects provided written informed consent before participation. Patient selection All adults with sNa concentration ≤ 127 mmol/l both on hospital admission and on the following day were identified through an automated laboratory system. This cut-off sNa value was selected because previous data from our cohort showed a significant upward inflection in inpatient mortality below that threshold [4]. Among these patients, only subjects who met all essential diagnostic criteria for SIAD, including euvolaemia, hyponatraemia and low serum osmolality with inappropriately raised urine osmolality and sodium, normal adrenocortical reserve, and exclusion of hypothyroidism [12, 13], participated in the study. Subjects were excluded if they met any of the following exclusion criteria: (1) aged < 18 years old, (2) presence of hypervolaemic hyponatraemia, (3) hypovolaemic hyponatraemia, (4) decompensated chronic liver disease, (5) decompensated heart failure, (6) renal impairment with In real-life clinical practice, the mainstay of SIAD treatment was fluid restriction in combination with discontinuation of offending drugs and treatment of underlying cause. When hyponatraemia was refractory to fluid restriction, some patients were referred to endocrinologists, usually after considerable delay, for consideration of pharmacological therapy. In addition to ‘standard’ clinical care, all patients had fu (...truncated)