Secondary Logo

Share this article on:

In Death, Truth Lies: Why Do Patients with Sepsis Die?

Tenhunen, Jyrki J. MD, PhD

doi: 10.1213/ane.0b013e3181a16554
Editorial: Editorials

From the Department of Critical Care Medicine, Tampere University Hospital, Tampere, Finland.

Address correspondence and reprint requests to Jyrki J. Tenhunen, MD, PhD, Department of Critical Care Medicine, Tampere University Hospital, PO Box 2000, 33521 Tampere, Finland. Address e-mail to

Yes, they still die. Yes, we still do our best not to let them die. And no, ultimately, we do not know why they die, do we?

It is no news that, even today, sepsis is associated with high mortality. All of us who regularly practice intensive care medicine have seen our patients with sepsis die. Yet, it seems acceptable to claim that we do not know ultimately why these patients die. Some die despite full continuing therapeutic efforts, although others die after therapy has been withheld or withdrawn. Most patients who die with or because of sepsis, die with established multiple organ dysfunction or failure.1–3 Although the clinical cause of death can be classified as “refractory septic shock,” “multiple organ failure,” or “acute circulatory failure,” the actual causes and mechanisms for treatment failure and death remain mostly unidentified.

Sepsis has perplexed clinicians for centuries. One of the few breakthroughs in understanding sepsis came from the Hungarian obstetrician Ignác Semmelweiss working in Austria. He was puzzled by the high mortality rate in one of the two obstetrics clinics in Vienna 150 yr ago. Although merely 3% of mothers died from childbed fever in the second clinic, mortality was some 18% in the first clinic of obstetrics. In the first clinic, it was customary for clinicians and students alike to perform autopsies (bare-handed) before examining the mothers during labor. After his colleague Dr. Kolletschka, a professor of pathology, died from sepsis (open wound in his finger during autopsy!), Semmelweiss realized, on reading through the autopsy report, that Kolletschka’s autopsy findings were identical with those of mothers dying from childbed fever.

“… suddenly a thought crossed my mind: childbed fever and the death of professor kolletschka were one and the same … his … sepsis and childbed fever must originate from the same source. [the cause] was to be found in the fingers and hands of students and doctors, soiled by recent dissections … [and they] … carry those death-dealing cadavers’ poisons into the genital organs of women in childbirth … the first principle is the absorption of the decomposed animal organic substance [by women in labor through the cadaveric particles from the hands of doctors and medical students] … as a result of this … there is … a change in composition of the blood.”4

In this historical case, postmortem findings in autopsy were a key element suggesting that contaminated hands infected mothers in the first clinic. Even though Semmelweiss tested his novel hypothesis immediately and reduced the mortality in the clinic to below 2%, his discovery was only later appreciated.

In this issue of Anesthesia & Analgesia, Austrian investigators have addressed postmortem findings in sepsis anew: Torgersen et al.5 report a reappraisal of the importance of postmortem investigation and findings in different organs in an attempt to elucidate mechanisms of death in patients dying with sepsis or septic shock. The authors painstakingly studied both clinical and postmortem records of all patients who died from sepsis or septic shock in their institution over a period of 10 yr. In almost 80% of the cases an unresolved infectious focus was found postmortem. Moreover, 90% of the patients with prolonged intensive care exceeding 7 days had persistent infectious foci. Only 52 of 97 autopsy-confirmed pneumonias were diagnosed during intensive care unit stay.

Previous autopsy studies in patients dying in intensive care units, suggest that the rate in discrepancies between pre- and postmortem diagnoses is highly variable,6–8 ranging from around 5% to 25% of diagnoses considered relevant to death; most of these are infections. Persistent or undiagnosed infections in dying patients are common. Although source control of infections should be emphasized,9,10 it is unclear to what extent these infectious foci, many of them undiagnosed pneumonias, contributed to the deaths.

Rather than supplying the reader and clinician with new insights into mechanisms of death in sepsis, Torgersen et al. offer confirmation of previous reports on inadequate premortem diagnoses. The reader should notice, however, the large number of autopsies performed in the authors’ institution, and thereby be alerted to the high credibility of the present investigation.

The authors quite correctly point out the limitations of this investigation. The uncontrolled, retrospective design of the study limits the conclusions. Furthermore, it is reasonable to assume that autolysis takes place soon after death. Thereby, some of the findings may be related to autolysis only. In addition, the macroscopic level of tissue-specific analyses clearly limits the value of the present report if the goal was to identify new mechanisms of death. The only way to overcome this problem would be to perform a prospective trial with early postmortem biopsies or autopsy comparing tissue-specific findings in septic patients to those of patients dying from other causes, such as rapidly progressing circulatory failure due to myocardial infarction. One such trial was reported just recently,11 with the limitation of a missing “control” group. Those authors also “only” report the histopathological findings in the liver based on immediate liver biopsy after death.

In which part of the report by Torgersen et al. is the truth? Where does it lie? Is unresolved infection and inadequate source control revealed? This is obviously sad but true. We cannot follow the 2000-yr-old rule by Galenos: “Ubi Pus, Ibi Evacua.” Does knowledge of the organ-specific macroscopic pathologies lead to new discoveries of mechanisms? Probably not. We can even argue that the postmortem findings lie at the organ level when time-related autolysis proceeds; we do not know if these findings are specific for sepsis. Furthermore, the functional changes of a specific organ are often greater than the macro- and microscopic findings observed in autopsy.12 In the present retrospective trial, most of the organs were indeed examined in an attempt to uncover new mechanisms; still, many unanswered questions remain. It may well be that more information on new mechanisms could be obtained if immediate postmortem tissue samples were available for analyses.

To be provocative, one could claim that most of the postmortem autopsy studies that focus on different organs have not provided researchers or clinicians with novel information on mechanisms of death, even though attempts have been made to carefully describe all the organs in death. However, in contrast to what Semmelweis proposed almost 150 yr ago, blood and its composition have rarely if ever been considered in this context. Is it possible that the composition and flow characteristics of blood change within the elastic pipelines, to the extent that function of multiple organs is compromised? As is the case in all postmortem (sepsis) studies, blood was not examined by Torgersen et al. Although much research in sepsis has focused on circulation and mediators in the blood, the physical characteristics of the blood have been largely overlooked. This is understandable. Blood changes the minute the sample is drawn, or the minute the patient dies; hence, research on fluid mechanic/fluid dynamic behavior of blood is challenging. Ironically, Semmelweiss talks about disintegration of blood, blood poisoning by definition. The fundamental question is whether blood poisons the tissue and organs, or alternatively, is itself literally poisoned, not only by bacteria, but by other bacterial or “animal” (= human?) material/particles.

Does looking at the composition of blood seem too far-fetched? There are several arguments in favor of taking a closer look at the composition of blood in sepsis. Blood, as a non-Newtonian fluid, does not have a constant viscosity.13,14 Blood flow can be turbulent,15 and non-Newtonian fluids with turbulent flow properties experience drag reduction (Toms’ effect16) (reviewed in Ref. 17). Minute (parts per million) concentrations of a high-molecular-weight (from 600 kD up) polymer in a fluid dramatically reduce friction resistance without changing the viscosity; examples of such polymers are DNA18 and hyaluronan, a structural component of vascular wall glycocalyx.19 Furthermore, bacteria produce drag-reducing polymers.20 High DNA concentrations in sepsis are associated with high mortality.21,22 Based on this, it can be hypothesized that DNA, hyaluronan and bacterial products induce drag reduction in blood during sepsis. Fluid mechanics of blood change: Toms’ effect changes the flow behavior (not viscosity) of blood, contributing to the characteristic hyperdynamic systemic circulation and low arterial blood pressure in sepsis. If this is the case, removing or cleaving high-molecular-weight polymers should restore normal rheological characteristics of blood.

To summarize: The article by Torgersen et al. re-emphasizes the importance of infectious source control in sepsis. This message from their article is clear, and provides once again an important reminder for the clinician. Actual new mechanisms leading to death, however, are NOT proposed in their study. To reveal new insights into the pathophysiology and treatment of sepsis, instead of focusing only on the inflammatory networks, mitochondrial alterations, or apoptotic events, maybe we should ask whether the blood indeed is “poisoned” in sepsis, as suggested in 1861 by Semmelweiss. Hypothetically, as a consequence, the composition and fluid dynamic nature of blood may be altered.

Back to Top | Article Outline


1. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–10
2. Alberti C, Brun-Buisson C, Burchardi H, Martin C, Goodman S, Artigas A, Sicignano A, Palazzo M, Moreno R, Boulmé R, Lepage E, Le Gall R. Epidemiology of sepsis and infection in ICU patients from an international multicentre cohort study. Intensive Care Med 2002;28:108–21
3. Varpula M, Karlsson S, Parviainen I, Ruokonen E, Pettilä V, Ala-Kokko TI, Kolho E, Rintala EM. Incidence, treatment, and outcome of severe sepsis in ICU-treated adults in Finland: the Finnsepsis study. Intensive Care Med 2007;33:435–43
4. Raju TN. Ignác Semmelweis and the etiology of fetal and neonatal sepsis. J Perinatol 1999;19:307–10
5. Torgersen C, Moser P, Luckner G, Mayr V, Jochberger S, Hasibeder WR, Dünser MW. Macroscopic post-mortem findings in 235 surgical intensive care patients with sepsis. Anesth Analg 2009;108:1841–7
6. Mort TC, Yeston NS. The relationship of pre mortem diagnoses and post mortem findings in a surgical intensive care unit. Crit Care Med 1999;27:299–303
7. Blosser SA, Zimmerman HE, Stauffer JL. Do autopsies of critically ill patients reveal important findings that were clinically undetected? Crit Care Med 1998;26:1332–6
8. Silfvast T, Takkunen O, Kolho E, Andersson LC, Rosenberg P. Characteristics of discrepancies between clinical and autopsy diagnoses in the intensive care unit: a 5-year review. Intensive Care Med 2003;29:321–4
9. Kumar A, Haery C, Paladugu B, Kumar A, Symeoneides S, Taiberg L, Osman J, Trenholme G, Opal SM, Goldfarb R, Parrillo JE. The duration of hypotension before the initiation of antibiotic treatment is a critical determinant of survival in a murine model of Escherichia coli septic shock: association with serum lactate and inflammatory cytokine levels. J Infect Dis 2006;193:251–8
10. Varpula M, Karlsson S, Parviainen I, Ruokonen E, Pettilä V; Finnsepsis Study Group. Community-acquired septic shock: early management and outcome in a nationwide study in Finland. Acta Anaesthesiol Scand 2007;51:1320–6
11. Koskinas J, Gomatos IP, Tiniakos DG, Memos N, Boutsikou M, Garatzioti A, Archimandritis A, Betrosian A. Liver histology in ICU patients dying from sepsis: a clinico-pathological study. World J Gastroenterol 2008;14:1389–93
12. Hotchkiss RS, Swanson PE, Freeman BD, Tinsley KW, Cobb JP, Matuschak GM, Buchman TG, Karl IE. Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med 1999;27:1230–51
13. Granger RA. Fluid mechanics. Mineola, NY: Dover Publications, 1995
14. Cokelet GR, Meiselman HJ. Macro- and micro-rheological properties of blood. In: Baskurt OK, Hardeman MR, Rampling MW, Meiselman HJ, eds. Handbook of Hemorheology and Hemodynamics. Fairfax, VA: IOS Press, 2007
15. Lee SE, Lee SW, Fischer PF, Bassiouny HS, Lothe F. Direct numerical simulation of transitional flow in a stenosed carotid bifurcation. J Biomech 2008;41:2551–61
16. Toms BA. Some observations on the flow of linear polymer solutions through straight tubes at large Reynolds number. Proceedings of the First International Rheology Congress 1948, Holland, North Holland Publishing Co., Amsterdam, 1949;2:135–41
17. Myagchenkov VA, Chichkanov SV. Toms Effect in Model and Real Systems. Russ J Appl Chem 2005;78:521–37
18. Hand JH, Williams MC. DNA and structural effects in turbulent drag reduction. Nature 1970;227:369–70
19. Thacker K, Kameneva M. (WO/2006/093957) Blood-soluble drag-reducing hyaluronic acid. A patent application in World Intellectual Property Organization. 2006.
20. Kenis PR. Drag reduction by bacterial metabolites. Nature 1968;217:240–2
21. Wijeratne S, Butt A, Burns S, Sherwood K, Boyd O, Swaminathan R. Cell-free plasma DNA as a prognostic marker in intensive treatment unit patients. Ann N Y Acad Sci 2004;1022:232–8
22. Saukkonen K, Lakkisto P, Pettilä V, Varpula M, Karlsson S, Ruokonen E, Pulkki K; Finnsepsis Study Group. Cell-free plasma DNA as a predictor of outcome in severe sepsis and septic shock. Clin Chem 2008;54:1000–7
© 2009 International Anesthesia Research Society