Results of severe sepsis treatment – two years of experience in a single centre
Danuta Gierek1, Małgorzata Kuczera1, Józefa Dąbek2, Dorota Piłat1, Aneta Kurtok-Nowak1
1Department of Anaesthesiology and Intensive Therapy, Teaching Hospital No 7, Medical University of Silesia in Katowice
2Department of Cardiology, Medical University of Silesia in Katowice
- Fig. 1. Organ failure in study groups
- Fig. 2. Bacteriological spectrum at the primary site of infection
- Fig.3. Antibiotics most commonly used
- Table 1. Demographic characteristics of patients
- Table 2. Places of severe sepsis recognition
- Table 3. Number of failed organs
- Table 4. Sources of severe sepsis
- Table 5. Therapy of patients with severe sepsis
- Table 6. APACHE II and TISS-28 scores, median of glucose concentration and use of septic bundles in severe sepsis patients
Background. Treatment of severe sepsis (SS) is a major challenge for an ITU, because of the high mortality. The severity of SS is scored according to the number of organ systems that have failed. We analysed the results of treatment of SS in the ITU of the Upper Silesian Medical Centre, and compared them to previously reported data obtained from other centres.
Methods. Between 2008 and 2009, 45 SS cases were treated. Demographic data, laboratory tests and therapeutic routines were analysed.
Results. The overall mortality rate was 58%. There was a positive correlation between mortality, male gender and advanced age. The most common initial sources of infection were the respiratory system and abdominal cavities. Serious metabolic disturbance and central nervous system failure markedly affected prognosis. In 56% of cases, five or more organ systems were affected.
Discussion. Survival in SS is related to the number of affected organ systems. All patients were admitted in critical condition requiring respiratory and inotropic support. Despite strict application of the Surviving Sepsis Campaign (SSC) recommendations and a relatively short time from diagnosis to commencement of adequate treatment, the mortality rate was found to be higher than the European average, but comparable to that from the Polish registry.
Conclusions. Advanced age, male gender and intraperitoneal pathology worsened the prognosis in SS. Mortality was directly related to the number of organ systems involved, and the number of coexistent diseases. An interdisciplinary approach during diagnosis and a reduced time to the start of intensive treatment are essential for survival. Ready access to updated databases on SS enables regular reviews of the results of treatment and improvement of algorithms.
Severe sepsis, defined as the systemic inflammatory response accompanied by multi-organ failure or tissue hypoperfusion, presents a therapeutic challenge. Despite the introduction of standards of management, severe sepsis remains the main cause of ITU deaths with the mortality rates ranging from 30 to 60%. During the last two decades, the number of deaths due to severe sepsis doubled; at present about 750 000 patients die annually because of severe sepsis. Its incidence is expected to rise, which is associated with new invasive techniques used and the higher number of immunocompromised patients. According to the Institute of Public Health, in the years 2003-2008, the prevalence of sepsis increased from 0.6 to 1.6 per 10 000 Polish citizens . It is estimated that patients diagnosed with severe sepsis constitute about 10-14% of ITU patients .
The data listed illustrate the social and medical implications of the disease and formed the basis for the activities undertaken by the Polish Working Group for Sepsis in 2001. The task of the Group is to collect information regarding patients with severe sepsis treated in ITUs using the Internet-based Notification Registry coordinated by the Surviving Sepsis Campaign. The guidelines for management of such patients and their use in practice improved significantly treatment outcomes and reduced the mortality by 16% in the years 2004-2007.
Due to high mortality rates and poor prognosis, the essential factor of successful therapeutic management is early diagnosis and multi-directional intensive therapy of severe sepsis [5, 6].
The objective of the study was to compare the selected parameters assessing the general condition of patients with severe sepsis and elements of management as well as to determine their effects on treatment outcomes.
The study involved patients hospitalized between 2008-2009 fulfilling the criteria of severe sepsis or septic shock according to the 1992 definition . Patients were allocated to two groups: group I – survivors; group II – non-survivors.
Age, gender, hospitalization time, mortality and places of sepsis recognition were assessed in both groups. Moreover, analysis included the number of failed organs, primary disease causing sepsis, etiological factor and coexisting diseases. The severity of patient`s condition on admission was evaluated according to APACHE II and TISS-28 scores. Furthermore, types of therapy were assessed: antibiotic treatment, mechanical lung ventilation, catecholamines, renal substitutive therapy, recombinant activated protein C (rhAPC) therapy, and surgery. The serum glucose levels were determined and the use of recommended septic bundles analysed .
Results were analysed statistically. Means and standard deviations were calculated. The significance of intergroup differences was checked using the t-test for independent samples. P <0.05 was considered significant.
In the period analysed, 646 patients were treated in ITU. In total, patients diagnosed with severe sepsis or septic shock constituted 7.0% (45): severe sepsis developed in 0.5% (3) whereas septic shock in 6.5% (42 patients). After therapy, 19 survivors (group I) were transferred to other wards (recovered from a life threatening condition); the remaining 26 (group II) died. The overall sepsis-related mortality in the period analysed was 57.78%, with 51.85% in 2008 and 66.67% in 2009.
The mean duration of treatment of all patients was 16±12.9 days: in group I – 14±9 days, and in group II – 17±14 days. Demographic data of patients are presented in Table 1.
In 28 patients severe sepsis was recognized before admission to ITU whereas in 17 sepsis developed during ITU hospitalization (Table 2.).
The time of transfer from emergency department to ITU was on average 15 min and did not significantly differ between the groups. Based on medical records, it was not possible to define the time of ITU admission from other wards.
The extent of multiorgan failure of patients is illustrated in Table 3. All patients developed respiratory failure, and 90% – circulatory insufficiency. The percentages of metabolic (64%), haemotologic (69%), renal (60%) and CNS (67%) failures were comparable. Liver dysfunction was least common (13%). In the group of non-survivors, the incidence of metabolic, renal and CNS failure was significantly higher. The insufficiency of the remaining organs was comparable in both groups (Fig. 1).
The most common causes of severe sepsis in patients admitted from the emergency department were found to be abdominal diseases whereas in those admitted from other wards – the respiratory dysfunction (Table 4).
Bacteriological tests were performed in all patients. In the majority of cases, mixed infections were found. In one case, isolated infection with Gram-positive bacteria was recognized. In 12 patients (27%) no microorganisms were cultured; positive cultures were found in 17 (38%): – 5 (26%) in group I whereas 12 (46%) in group II. Gram-positive bacteria were detected in 30% of cases and Gram-negative ones in 70% of patients. The coexisting fungal infections were diagnosed in 24% of patients.
The most common bacteria cultured from the respiratory system were Acinetobacter baumannii – in group I and Klebsiella pneumoniae – in group II. In the abdominal cultures, the most common pathogens were Pseudomonas aeruginosa and Acinetobacter baumannii – in group I; in group II, none of the pathogens dominated. The microorganisms cultured from primary sites of infection are given in Fig. 2.
The coexisting diseases were diagnosed in 89% of patients. In group I, systemic connective tissue disease (9), arterial hypertension (8) and chronic renal failure (5) were relatively common. Neoplastic disease was earlier diagnosed in one patient and chronic obstructive pulmonary disease in 2 patients. In group II, patients suffered mainly from cardiovascular diseases (18), including ischaemic heart disease (12), atrial fibrillation (7), arterial hypertension (8), septal defects (4). Moreover, obliterative arteritis (3), pancreatic disease (5), diabetes mellitus (7), and chronic renal failure (11) were detected.
The most common antibiotic used in both groups was imipenem/cilastine. The remaining antibiotics administered during therapy are listed in Fig. 3.
All patients were mechanically ventilated; in group I – for 8.5 days and in group II – for 16.3 days, on average. Vasoconstricting drugs were administered to 89% of patients, most commonly noradrenaline (group I – 78.9%, group II – 88.5% of patients), followed by dopamine (36.8% and 53.8%) and dobutrex (5.3% and 11.5%). Adrenaline was used in 46.2% of group II patients. The drugs were administered over 4.3 days in group I and 13.9 days in group II.
In all septic shock patients, small doses of steroids were used. Non-fractionated and small-molecular-weight heparin was prophylactically administered in 87% of cases, activated protein C – in 6.7%. Surgical procedures were performed in 36% of patients. Non-survivors were more frequently operated on compared to survivors (38% and 32%, respectively). Likewise, haemofiltration was more frequently used in non-survivors; survivors were more often haemodialysed (Table 5).
General status of patients on admission assessed according to APACHE II score was significantly better in survivors vs non-survivors. The TISS assessment was also found significantly different (Table 6).
The median of glucose concentration during the first 24 h after diagnosis was 148.6±45.14 mg dL-1 and was significantly different between non-survivors (152.6 mg dL-1) and survivors (143.2 mg dL-1) (Table 6).
The efficacy of severe sepsis therapy was evaluated according to the use of septic bundles and was found comparable in both groups of patients (I – 79.5%, II – 77.3%). The difference was observed in the percentage use of septic bundle procedures in 2008 vs 2009. In group I, the use of resuscitation bundles increased in the period analysed by 15.38% in group II – by 11.3%. Moreover, in the years 2008/2009 the use of elements of therapeutic bundles increased by 9.3% in group I and by 3.3% in group II (Table 6).
The incidence of severe sepsis in our study (7%) was twofold lower than in the countrywide studies yet similar to the results from the Lódz centre (8.9%) [7, 8]. The difference is likely to result from the specificity of our unit, which admits a high proportion of patients after extensive vascular procedures, in haemorrhagic shock, and with serious cardiovascular diseases.
The mean age of our patients was slightly lower whereas the gender profile similar to the country- and world-wide populations [9, 10, 11, 12]. The mean treatment time in ITU was by 3 days shorter (16 vs 19) and mortality (57%) comparable to the national mean (55%).
It is worth stressing that, contrary to the national data, a high percentage of patients with severe sepsis was admitted to our unit directly from the emergency department (36.7%). In the national registry, this percentage was 8.5%, in the Lódz centre – 18.5% [7, 8]. Moreover, the shorter time between diagnosis and ITU admission enabled prompt use of septic bundles.
Analysis of the number of failed organs showed significantly increased mortality at 5 or more failed organs (76%). According to the national data, mortality increases already at 4 failed systems and is estimated at 60.1% . In the group of non-survivors, the incidence of CNS, metabolic and renal failures was higher. It is well known that mortality of patients with severe renal failure increases when accompanied by the failure of other organs. Furthermore, in septic patients, CNS failure is an independent risk factor of death .
Contrary to the national data, in our material the primary site of pathology leading to the development of sepsis in survivors was the respiratory system (53% vs 28%), followed by the abdominal cavity (32% vs 47%). In non-survivors, the data were more comparable with the national estimates (abdominal cavity– 39%, respiratory system – 39%). Our results are in accordance with those from the USA and European countries [9, 10, 11, 12]. In the majority of patients arriving to ITU from the emergency department, the primary cause of sepsis was abdominal diseases, which is consistent with the Polish data [7, 8].
The majority of infections were caused by Gram-negative bacteria; in our material – 70% vs 48.2% countrywide and 67% in the Lódz centre. The percentage of fungal infections was comparable to other centres. The number of positive blood cultures was slightly lower than the national data (38% vs 45%). In non-survivors, the incidence of bacteraemia was higher, which confirms its negative effects on prognosis. Acinetobacter baumannii was the pathogen most commonly cultured from the bronchial lavage. This pathogen is often responsible for colonization but not infection. Klebsiella pneumoniae, on the other hand, mainly cultured from patients with diagnosed infections , is resistant to the majority of antibiotics, including carbapenems. Therefore, in our study it was most commonly detected in the respiratory cultures of non-survivors.
Moreover, 9 patients from group I had connective tissue diseases, which was associated with special surveillance during therapy (immunocompromised patients and increased risk of infections). Thanks to early diagnosis and prompt transfer of patients to ITU, 78% of them survived severe sepsis. Cardiovascular and metabolic diseases were more frequently observed in non-survivors.
Due to multi-organ failure in various configurations, patients required individualized treatment protocols. Renal substitutive therapy was used in patients with renal failure according to RIFFLE criteria . Continuous veno-venous haemofiltration was more common in the group with advanced multi-organ failure; nonetheless beneficial effects on treatment outcomes were not observed . Haemofiltration was more commonly used in non-survivors. This may indicate that this method was introduced too late in advanced multi-organ failure and positive treatment outcomes were infeasible [15, 16].
In the analysed group, rhAPC was used in three cases; in two of them therapy failed. Patients with rhAPC therapy (non-survivors) were admitted from other hospitals due to earlier multi-organ failure and thus the rhAPC treatment was delayed. According to the recommendations, the therapy should be used as early as possible in patients at high risk of death and with moderately increased multi-organ failure .
Heparin was used for antithrombotic prophylaxis; clotting disturbances were normalized by transfusion of fresh frozen plasma and platelets. According to Rybicki , the efficacy of heparin in the treatment of developed clotting disorders in such a group of patients was not demonstrated.
In our study, the number of patients undergoing surgery was slightly lower compared to averaged national results (36% vs 48%), which probably resulted from the dominating profile (internal diseases) of patients treated for severe sepsis in our ITU.
Despite severe conditions of patients, their TISS score was not high and treatment costs did not correspond to the National Health Funds funding.
Aggressive insulin therapy was not administered within the first 24 h after severe sepsis diagnosis due to possible accidental hypoglycaemia, particularly in unconscious patients. The median of blood glucose concentration during the first treatment day in survivors was within upper limits whereas among non-survivors exceeded the acceptable values, which confirms the prognostic value of this parameter for sepsis . In accordance with the recommendations, the blood concentration of glucose during treatment of sepsis should not exceed 150 mg dL-1 .
The findings demonstrated progress in the use of elements of septic bundles in our centre during the last two years. Moreover, the elements of resuscitation and therapeutic bundles were also more commonly used; unfortunately, this did not translate into lower mortality rates.
The incidence of renal and metabolic failure, consciousness disorders, APACHE II score >25 and dysfunction of 5 or more systems was higher among non-survivors. The most likely cause of high mortality in patients with severe sepsis in ITU is delayed referral to the appropriate reference centres.
Internet-based registration of cases of ITU severe sepsis evidences the magnitude of the problem in Poland . Averaged national data are used to compare treatment outcomes in individual intensive therapy units and advances in early diagnosis and treatment of severe sepsis can be monitored and adjusted to new guidelines and specificity of patients admitted to ITU, with the hospital profile considered.
1. Thanks to quick transfer of patients with severe sepsis from the emergency department to intensive therapy unit, adequate therapy can be administered in optimal time and irreversible multi-organ failure avoided.
2. The most common sources of infections in patients with severe sepsis are the lungs and abdominal cavity.
3. Metabolic, CNS, and renal failures, APACHE II >25, bacteraemia, insufficiency of 5 or more organs are the factors predicting poor prognosis.
4. Wide use of procedures included in septic bundles is likely to reduce mortality rates in patients with severe sepsis.
1. Kübler A: Sepsa – definicje i epidemiologia. Med Sci Rev 2005; 1: 5-10.
2. Narodowy Instytut Zdrowia Publicznego, http://www.pzh.gov.pl.
3. Kübler A, Mayzner-Zawadzka E, Durek G, Gaszyński W, Karpel E, Mikaszewska-Sokolewicz M, Majak P: Results of severe sepsis treatment program using recombinant human activated protein C in Poland. Med Sci Monit 2006; 12: CR107-1125.
4. Dellinger R, Levy M, Carlet J, Bion J, Parker M, Jaeschke R, Reinheart C, Angus D, Brun-Buisson C, Beale R, Calandra T, Dhalnaut J, Gerlach H, Harvey M, Marini J, Marshall J, Ranieri P, Ramsay G, Sevransky J, Thompson B, Townsend S, Vender J, Zimmerman J, Vincent J: Międzynarodowe wytyczne postępowania w ciężkiej sepsie i wstrząsie septycznym – Surviving Sepsis Campaign 2008. Medycyna Praktyczna 2008; 4, 96-109.
5. Slade E, Tamber P, Vincent J: The Surviving Sepsis Campaign: Raising awareness to reduce mortality. Crit Care 2003; 7: 1-2.
6. Jaeschke R, Brożek J, Dellinger P: Aktualizacja międzynarodowych zaleceń dotyczących postępowania w ciężkiej sepsie i wstrząsie septycznym 2008 – czy powinniśmy zmienić naszą praktykę kliniczną? Medycyna Praktyczna 2008; 4, 27-31.
7. Kübler A, Durek G, Zamirowska A, Duszyńska W, Palysińska B, Gaszyńska W, Pluta A: Severe sepsis in Poland – results of internet surveillance of 1043 cases. Med Sci Monit 2004; 10: CR635-641.
8. Glapiński A, Jaszczuk E, Gaszyński W: Ciężka sepsa w materiale Oddziału Intensywnej Terapii Wojewódzkiego Szpitala Specjalistycznego im. M. Kopernika w Łodzi w latach 2001-2004. Sepsis 2008; 1: 5-10.
9. Angus D, Linde-Zwirble W, Lidicker J, Clermont G, Carcillo J, Pinsky M: Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated cost of care. Crit Care Med 2001; 29: 1303-1309.
10. Martin G, Brunkhorst F, Janes J, Reinhart K, Sundin D, Garnett K, Beale R: The international PROGRESS registry of patients with severe sepsis: drotrecogin alfa (activated) use and patient outcomes. Crit Care 2009; 13: R103.
11. Blanco J, Muriel-Bombin A, Sagredo V, Taboada F, Gandia F, Tamayo L, Collado J, Garcia-Labattut A, Carriedo D, Valledor M, De Frutos M, Lopez M, Caballero A, Guerra J, Alvarez B, Mayo A, Villar J: Incidence, organ dysfunction and mortality in severe sepsis: a Spanish multicenter study. Crit Care 2008; 12: R158.
12. Brun-Buisson C, Meshaka P, Pinton P, Vallet B: EPISEPSIS: a reappraisal of the epidemiology and outcome of severe sepsis in French intensive care units. Inten Care Med 2004; 30: 580-588.
13. Rybicki Z: Sepsa, wstrząs septyczny; w: Intensywna terapia dorosłych, (Ed.: Rybicki Z) Mak Med, Lublin, 2009: 453-454, 458.
14. Wójkowska-Mach J, Baran M, Drwiła R, Foryciarz E, Misiewska-Kaczur A, Romaniszyn D, Heczko PB: Etiologia i lekooporność zapalenia płuc u chorych po operacjach z zakresu chirurgii serca leczonych respiratorem. Anaesthesiol Intensive Ther 2009; 41: 224-229.
15. Ricci Z, Cruz D, Ronco C: The RIFLE criteria and mortality in acute kidney injury. A systematic review. Kidney Int 2008; 73: 538–546.
16. Page B, Vieillard-Baron A, Chergui K, Peyrouset O, Rabiller A, Beauchet A, e Aegerter P, Jardin F: Early veno-venous haemodiafiltration for sepsis-related multiple organ failure. Crit Care 2005; 9: R755-R763.
17. Barteczko B, Zamirowska A, Kubler A: Stężenie glukozy we krwi chorych z ciężką sepsą i wstrząsem septycznym. Anestezjol Inten Terap 2007; 3: 152-155.
Oddział Anestezjologii i Intensywnej Terapii
SPSK nr 7 ŚUM, GCM
ul. Ziołowa 45/47, 40-635 Katowice-Ochojec
tel. 032/359 81 00