Advertisement

Sodium: sign, signifier, or signified, of sepsis?

  • Mai O'Sullivan
    Affiliations
    Department of Anaesthesia & Critical Care, Children's Health Ireland at Crumlin, Dublin, Ireland
    Search for articles by this author
  • Kevin F McCarthy
    Correspondence
    Corresponding Author: Kevin F McCarthy, Discipline of Paediatrics, School of Medicine, Trinity College Dublin, Ireland.
    Affiliations
    Department of Anaesthesia & Critical Care, Children's Health Ireland at Crumlin, Dublin, Ireland

    Discipline of Paediatrics, School of Medicine, Trinity College Dublin, Ireland
    Search for articles by this author
Published:December 16, 2020DOI:https://doi.org/10.1016/j.ejim.2020.12.002
      85% of the known universe is made up of something we cannot directly observe as it does not absorb, emit, or reflect light. We can only infer the existence of dark matter through its effects on gravitational waves and because, otherwise, galaxies would fly apart, and since this has not happened (yet), there must be something holding them together. With current technology, any proxy measures to infer its existence are at least two degrees of separation removed from the substance of actual interest: dark matter bends gravity, gravity bends light and it is in that bending of light that we try to pick up signal from the bewildering background noise of a vast and expanding universe. And so it must also seem in trying to identify simple, reliable, and accurate predictors of outcome when medical knowledge is expanding exponentially in complexity and volume. This rapid rate of change means that medicine may now have evolved beyond the point where any one human mind can comprehend it, which has led to interest in augmenting clinical gestalt with the likes of machine learning [
      • Obermeyer Z
      • Lee TH.
      Lost in Thought - The Limits of the Human Mind and the Future of Medicine.
      ] But it is not just that medical knowledge is becoming more complex, so are the patients, healthcare systems, and the world in which we live. It feels like an understatement to be referencing the, originally military, terminology of “VUCA” (short for Volatility, Uncertainty, Complexity, and Ambiguity) [
      • Bennett N
      • Lemoine J.
      What VUCA Really Means for You.
      ] in November 2020, 10 months into a global pandemic, but it is apposite nonetheless: global societal disruption, an unknown timeline for a vaccine, the possibility of short-lived immunity, the rise and fall of various touted cures in several short months, from hydroxychloroquine to dexamethasone.
      However, the arrival of Covid-19 has only accelerated and amplified trends that were already present in medicine. Paradigms were not just shifting but inverting: reduced mortality became increased multiple co-morbidities; centres of excellence became ‘hospital at home’; the intersection of (by necessity) generalist clinicians with expert patients. And so to sodium and sepsis; not only a hardworking monovalent cation but also potentially a prognostic biomarker hiding in plain sight? The attraction of serum sodium as such a biomarker is obvious; it is simple, easily and routinely checked, with the possibility of point-of-care testing in emergency departments. So it would seem, in a series of recent reports in the European Journal of Internal Medicine, including that by Castello et al., in this issue [
      • Castello LM
      • Gavelli F
      • Baldrighi M
      • Salmi L
      • Mearelli F
      • Fiotti N
      • Patrucco F
      • Bellan M
      • Sainaghi PP
      • Ronzoni G
      • Di Somma S
      • Lupia E
      • Muiesan ML
      • Biolo G
      • Avanzi GC
      Hypernatremia and moderate-to-severe hyponatremia are independent predictors of mortality in septic patients at emergency department presentation: A sub-group analysis of the need-speed trial.
      ]. This was a subgroup analysis of the Need-Speed trial, a multicentre observational study that enrolled patients across five Italian hospitals. They identified dysnatremia, including both hypo- and hyper-natremia, in the emergency department to be a predictor of inpatient mortality at both seven and 30 days, albeit with less significance at 30 days. Those at highest risk were found to be those with moderate-to-severe hyponatremia and hypernatremia.
      This appears to align with other recently published reports. Muller et al [
      • Müller M
      • Schefold JC
      • Guignard V
      • Exadaktylos AK
      • Pfortmueller CA
      Hyponatraemia is independently associated with in-hospital mortality in patients with pneumonia.
      ] in Bern, Switzerland found that hyponatremia on hospital admission was an independent risk factor for in-hospital mortality in 610 patients diagnosed with pneumonia, irrespective of confounding factors. However it was only associated with intensive care (ICU) admission when disease severity was not taken into account. In addition to this, Potasso et al [
      • Potasso L
      • Sailer CO
      • Blum CA
      • Cesana-Nigro N
      • Schuetz P
      • Mueller B
      • Christ-Crain M
      Mild to moderate hyponatremia at discharge is associated with increased risk of recurrence in patients with community-acquired pneumonia.
      ]. in Basel, Switzerland, reported on 725 patients admitted with pneumonia and found that patients with hyponatremia at discharge have a higher recurrence rate of pneumonia within the first 180 days after hospitalization when compared to patients with normal serum sodium levels at discharge. Moreover, this this association was particularly strong for patients who were hyponatremic both on admission and at discharge. Therefore it would seem that hyponatremia on admission has prognostic value extending beyond hospital admission to alter recurrence six months later.
      Hyponatremia is common and already a part of prognostic scoring systems such as the Pneumonia Severity Index (PSI) [
      • Fine MJ
      • Auble TE
      • Yealy DM
      • Hanusa BH
      • Weissfeld LA
      • Singer DE
      • Coley CM
      • Marrie TJ
      • Kapoor WN
      A prediction rule to identify low-risk patients with community-acquired pneumonia.
      ] and the APACHE II [
      • Knaus WA
      • Draper EA
      • Wagner DP
      • Zimmerman JE
      APACHE II: a severity of disease classification system.
      ] score. Could it have been doing the heavy lifting within those scoring systems all along? Attractive as that possibility might be, as with so much else in medicine, the reality is probably more ambiguous and comes with several caveats attached. Which brings us to McCarthy et al, and their 16-year review of 106,586 episodes of medical emergencies in 54,928 unique patients admitted to a tertiary hospital in Dublin, Ireland [
      • McCarthy K
      • Conway R
      • Byrne D
      • Cournane S
      • O'Riordan D
      • Silke B
      Hyponatraemia during an emergency medical admission as a marker of illness severity & case complexity.
      ]. While they found hyponatremia to be strong independent predictor of both mortality and length of stay, this was co-founded by the fact that it more frequently affected older patients who had worse acute illness scores and more complex presentations. The other finding over the study period, was a reduction in the frequency of hyponatremia and mortality; in 2002-09, 63% of patients had a sodium level in the lowest decile and the 30-day in-hospital mortality was 11.2%, compared with 2010-17, when 48% of patients had a sodium level in the lowest decile and 30-day in-hospital mortality decreased to 7.4%.
      This highlights the ‘VUCA’ principle of the applicability of scoring systems in populations that are also rapidly changing themselves. Not only is it a case of ‘the map is not the territory’, but that we are attempting cartography in a landscape that is itself changing. While sodium derangements indicate the need for greater clinical attention, fixating on the minutiae of sodium homeostasis may be placing emphasis on the signifier not the signified. However, neither are changes in sodium an epiphenomenon or accidental finding [
      • Obradović D
      • Esquinas AM.
      Hyponatremia at discharge: A solid risk or accidental findings in Community-acquired pneumonia.
      ]. Hyponatremia is likely a facet of a larger whole; it is a signifier, but is not separate or distinct from that which it is signifying.
      Based on these recent findings, while hyponatremia at presentation in sepsis may have utility as a biomarker of inpatient mortality and if still present at discharge, recurrence of pneumonia, the appropriate interventions may not be a unimodal focus on correction of sodium levels, but rather allocation to a higher risk category that mandates additional bundles of complex care that include frequent monitoring, multidisciplinary input and more intensive follow-up. Beyond respiratory sepsis, the compromise of accuracy versus usability of a model or biomarker may unfortunately also apply here. This is described as Bonini's Paradox and has been famously and succinctly stated by the French poet and philosopher, Paul Valéry: "Everything simple is false. Everything which is complex is unusable." [
      • Valéry Paul
      Collected Works of Paul Valéry.
      ]

      Declaration of Competing Interest

      The authors have no conflicts of interest to declare.

      References

        • Obermeyer Z
        • Lee TH.
        Lost in Thought - The Limits of the Human Mind and the Future of Medicine.
        N Engl J Med. 2017 Sep 28; 377 (PMID: 28953443; PMCID: PMC5754014): 1209-1211https://doi.org/10.1056/NEJMp1705348
        • Bennett N
        • Lemoine J.
        What VUCA Really Means for You.
        Harvard Business Review. 2014; 92 (2014, Available at SSRN:)
        • Castello LM
        • Gavelli F
        • Baldrighi M
        • Salmi L
        • Mearelli F
        • Fiotti N
        • Patrucco F
        • Bellan M
        • Sainaghi PP
        • Ronzoni G
        • Di Somma S
        • Lupia E
        • Muiesan ML
        • Biolo G
        • Avanzi GC
        Hypernatremia and moderate-to-severe hyponatremia are independent predictors of mortality in septic patients at emergency department presentation: A sub-group analysis of the need-speed trial.
        Eur J Intern Med. 2020;
        • Müller M
        • Schefold JC
        • Guignard V
        • Exadaktylos AK
        • Pfortmueller CA
        Hyponatraemia is independently associated with in-hospital mortality in patients with pneumonia.
        Eur J Intern Med. 2018; 54 (Aug. Epub 2018 Apr 13. PMID: 29657106): 46-52https://doi.org/10.1016/j.ejim.2018.04.008
        • Potasso L
        • Sailer CO
        • Blum CA
        • Cesana-Nigro N
        • Schuetz P
        • Mueller B
        • Christ-Crain M
        Mild to moderate hyponatremia at discharge is associated with increased risk of recurrence in patients with community-acquired pneumonia.
        Eur J Intern Med. 2020; 75 (May, Epub 2020 Jan 15. PMID: 31952985): 44-49https://doi.org/10.1016/j.ejim.2019.12.009
        • Fine MJ
        • Auble TE
        • Yealy DM
        • Hanusa BH
        • Weissfeld LA
        • Singer DE
        • Coley CM
        • Marrie TJ
        • Kapoor WN
        A prediction rule to identify low-risk patients with community-acquired pneumonia.
        N Engl J Med. 1997 Jan 23; 336 (PMID: 8995086): 243-250https://doi.org/10.1056/NEJM199701233360402
        • Knaus WA
        • Draper EA
        • Wagner DP
        • Zimmerman JE
        APACHE II: a severity of disease classification system.
        Crit Care Med. 1985; 13 (Oct. PMID: 3928249): 818-829
        • McCarthy K
        • Conway R
        • Byrne D
        • Cournane S
        • O'Riordan D
        • Silke B
        Hyponatraemia during an emergency medical admission as a marker of illness severity & case complexity.
        Eur J Intern Med. 2019; 59 (Jan. Epub 2018 Aug 7. PMID: 30097216): 60-64https://doi.org/10.1016/j.ejim.2018.08.002
        • Obradović D
        • Esquinas AM.
        Hyponatremia at discharge: A solid risk or accidental findings in Community-acquired pneumonia.
        Eur J Intern Med. 2020; 78 (Aug. Epub 2020 Jun 11. PMID: 32536563): 135-136https://doi.org/10.1016/j.ejim.2020.04.064
        • Valéry Paul
        Collected Works of Paul Valéry.
        Analects. Translated by Stuart Gilbert. 14. Princeton University Press, 1970: 466