Impact of multimodal strategy on sepsis bundle compliance and in- hospital mortality in patients with hematologic malignancies

doi:10.21203/rs.3.rs-2953924/v1

Introduction: Sepsis is one of the leading causes of mortality in patients with hematologic malignancies. Our aim is evaluate the impact of multimodal strategy on sepsis bundle compliance and in-hospital mortality in patients with hematologic malignancies.

Methods:We conducted a quasi-experimental study that included patients with hematologic malignancies and sepsis or septic shock, between January 1, 2017 and August 31, 2020. The study comprised two phases: phase 1, survey of sepsis and sepsis shock cases and reformulation of the sepsis team and sepsis bundle; phase 2, staff training, monthly meetings of the sepsis team and feedback for care teams. The followup of patients was for 30 days, during hospitalization, or until death. Data were analyzed with descriptive and inferential statistics, with a 95% significance interval. And the tests considered a 5% significance level.

Results: A total of 329 (42.5%) patients were included. The proportion of sepsis was 241 (73.3%) and septic shock 88 (26.7%). We observed an increase in the 3h-bundle compliance in the phase 2 (86.8%), compared to the phase 1 (71.3%), with p=0.001. The in-hospital mortality was lower in the 2nd phase (38.2%), compared to the first phase (51.5%), with p=0.024. These results were mainly observed in cases of sepsis identified in the wards.

Conclusions:The multimodal strategy had an impact on increasing the rate of compliance to the sepsis bundle. In-hospital mortality significantly reduced in the intervention phase of the study.

Sepsis

Bundle compliance

Multimodal strategy

Hematologic malignancies

In-hospital mortality

Infection is a frequent complication in patients with hematologic malignancies. This is mainly associated with interventions for the treatment of underlying disease, such as cytotoxic chemotherapies, use of invasive devices, like central venous catheter, neutropenia and mucositis [1]. Therefore, this is a population at high risk of sepsis, an event associated with high morbidity and mortality.

Brazil has a high mortality rate for sepsis, varying in different regions of the country: Southeast (51.2%), Midwest (70%), Northeast (58.3%), South (57.8%) and North (57.4%) [2]. Well-defined management bundles are of fundamental importance to reduce this mortality rate. Ferrer R et al., 2009, demonstrated that the mortality decreased from 58.7% in 1998–2003 to 43% in 2004–2008 after the introduction of sepsis bundle [3].

The Surviving Sepsis Campaign Guidelines (2016) recommends the 1-hour bundle that includes lactate collection, blood culture collection, antibiotics within the first hour, and volume replacement if lactate greater than 4 mmol or hypotension. These interventions ensure that patients will receive guided treatment according to their symptoms [4].

Empiric treatment with broad spectrum antibiotics should be started immediately in neutropenic patients with sepsis. Study including over 2,000 patients during sepsis and hypotension who received effective antimicrobial therapy within the first hour showed increased survival. Conversely, it was observed that for every hour of delay in antimicrobial administration over the course of 6 hours, a 7.6% decrease in survival was observed [5].

Due to the severity of these patients with hematologic malignancies the implementation of multimodal strategies can improve the management of sepsis bundle and decrease mortality. A multicenter study conducted in medical surgical ICUs across Spain based on the surviving sepsis campaign guidelines through an educational program for improvement in the care process and outcome for sepsis. Training was given to medical and nursing staff in the emergency department, wards and ICU as defined for the management of sepsis according to the guidelines. Included were 854 patients pre-intervention, 1456 post-intervention, and 247 at one-year follow-up after the intervention in a subset of 23 ICUs. Patients in the post-intervention cohort had lower in-hospital mortality (44.0% vs. 39.7%; P = 0.04). Compliance to care process variables also improved after the intervention in the sepsis resuscitation (5.3% vs. 10.0%; P < 0.001) and the sepsis management bundle (10.9% vs. 15.7%; P = 0.001). An educational program to promote sepsis and septic shock care bundles is related to lower mortality and better guideline achievement [6].

In the present study, we aim to evaluate the competence of a multimodal program for improving sepsis bundle compliance and reducing 30-day mortality from sepsis in patients with hematologic malignancies. Despite the benefits observed in these studies, few studies focus on multimodal strategy in sepsis bundle compliance.

We conducted a quasi-experimental study, which consecutively included patients with hematological neoplasms and a diagnosis of sepsis or septic shock, in the period between January 1, 2017 and August 31, 2020. It was performed in a tertiary hospital specialized in the care of hematological neoplasms patients, in the city of Sao Paulo, Brazil. Patients in palliative care, under 18 years of age, and confirmed SARS-CoV-2 diagnosis were excluded from the study.

Intervention

The study comprised two phases: observational period, intervention and post-intervention. In the phase 1 (January 1, 2017 to December 31, 2018), the survey of sepsis and sepsis shock cases was performed with analysis of the interval of sepsis bundle dysfunction and opening, bundle compliance, and in-hospital mortality rates compared to previous periods. At the end of this phase, there was a reformulation of the sepsis team, with the inclusion of board members and strategic physicians, nurses, and pharmacists from each interface. The sepsis bundle was revised according to changes in the definitions of sepsis and septic shock, and also the empirical antimicrobial therapy bundle. Based on these changes, during the phase 2 (January 1, 2019 to August 31st, 2020), didactic and practical trainings sessions were realized. The training was carried out by qualified professionals from the Hospital Infection Control Service (SCIH), allowing for uniformity of conduct. A pre and post-intervention test was applied as a form of learning evaluation. In conjunction with the training, posters were distributed throughout the wards and ICUs. The sepsis team started monthly meetings to discuss process and result indicators, goals, and new training when necessary. Additionally, indicator feedbacks for care team through reports and on-site discussions were initiated monthly. After the initial training period, the sepsis team meetings were held to discuss indicators and actions to address the results. Feedback to the care teams was also maintained through monthly reports and discussion of cases. When new professionals were admitted, didactic and practical training on the sepsis bundle was conducted.

In the study, patients were followed up for 30 days during hospitalization or until death. Data were extracted from the electronic medical records.

Outcomes

We evaluated two main outcomes in the study, compliance to the sepsis bundle and 30-day and in-hospital mortality. The 3-hour sepsis bundle included the following measures: (a) serum lactate measurement; (b) blood cultures collected before antibiotic initiation and culture from all suspected infection sites with time recorded in the medical record; (c) administration of broad-spectrum intravenous antibiotic therapy within the first hour, time recorded in the prescription; (d) aggressive volume replacement (at least 30 ml/kg/h of crystalloids) in patients who had altered lactate or hypotension. Bundle compliance was considered when all four components were covered in patient care. Other evaluated components were: (e) vasopressor within 6 hours if persistent hypotension, (f) Repeat measurement of lactate if previous altered lactate, and (g) assessment of therapeutic response through clinical and hemodynamic parameters of the patient, with appropriate medical record.

Definitions

At the study hospital, Modified Early Warning Score (MEWS) was used for initial screening of patients with sepsis (score ≥ 4). Sepsis and septic shock were defined according to the criteria of the Surviving Sepsis Campaign [7]. The severity classification of patients was done through Sequential Organ Failure Assessment (SOFA) at the time of sepsis diagnosis.

Neutropenia was considered as absolute neutrophil count less than 500 cells per mm3 on at least 2 consecutive days at the time of sepsis identification. Severe neutropenia was defined as an absolute neutrophil count of less than 100 cells per mm3 on at least 2 consecutive days. The focus of infection was defined according to 2020 CDC/NHSN criteria [8].

Statistical analysis

The analyses were performed with the aid of R 4.0.2 software (RCore Team, 2020) [9]. For the tests, a significance level of 5% was considered. As the model had numerous variables, we used the stepwise backward selection model according to Akaike's criterion (AIC) (Hosmer and Lemeshow, 2002) [10]. The data were presented with the respective odds ratio estimates with 95% confidence interval.

To analyze the 30-day and in-hospital outcome, logistic regression models were adjusted, considering as explanatory variables besides the phase, the main variable of interest, other variables associated or possibly associated with the outcome (death), based mainly on the results of univariate analysis. The results of the complete models (pre-selected variables) and by selection using the stepwise backward method according to Akaike's criterion (AIC) (Hosmer and Lemeshow, 2002) are presented, keeping the period variable, even if not significant. The results were presented with the respective odds ratio estimates with a 95% confidence interval [10].

During the study period, 775 patients presented with sepsis criteria. Of these, 329 (42.5%) had hematologic malignancy diagnosis. The mean age of the population was 58.6 years, and females were the most prevalent (50.2%). Among the baseline diagnoses, lymphoma was the most frequent (24.3%), followed by multiple myeloma (19.8%), and acute leukemia (16.1%). In the 2nd period, we observed a lower rate of acute leukemia, and neutropenia at diagnosis of sepsis. The rate of patients undergoing chemotherapy at the diagnosis of sepsis was higher in phase 2 of the study, as shown in Table 1. Among BMT recipients, we observed a higher frequency of autologous in phase 2 of the study (p=0.038).

The diagnosis of sepsis occurred in 241 (73.3%) and septic shock in 88 (26.7%) of the patients, with no difference between the two phases. The main infection sources were pulmonary (52.0%), abdominal (17.6%), central venous catheter-associated bloodstream (7.6%), and urinary tract 16 (4.9%). The 30-day and in-hospital mortality was 38.3% and 42.2%, respectively.

Table 1

Sepsis bundle compliance

Overall compliance to the sepsis bundle in the study was 82.1% (n = 270). During the phase 2, we observed a reduction in the time between the first detected organic dysfunction and recognition of sepsis, with a statistically significant difference (p = 0.012). In the 3-hour bundle, all components showed an increase in the compliance rate, but with statistical difference only antibiotic within 1 hour (p = 0.010) and blood cultures before starting antibiotic therapy (p<0.001). Comparison of compliance to all sepsis bundle components in both phases of the study is described in Table 2. We only observed a trend towards increased compliance to the complete sepsis bundle in phase 2 (p = 0.059).

Table 2

Compliance to the sepsis bundle was evaluated in different care and hospitalization departments: emergency department, wards and intensive care units. We observed increased compliance to the sepsis bundle in the wards and ICU in the 2nd phase, but with a statistically significant difference only in the wards (p=0.023).

Figure 1

30 day and in-hospital mortality

We observed a reduction in 30-day mortality from 42.6% in the 1st phase to 36.4% in the 2nd phase, but with no statistically significant difference, p=0.288. However, in-hospital mortality reduced from 51.5% to 38.2%, with p=0.024. As demonstrated in Figure 2, in-hospital mortality occurred in the wards, with p=0.017. No significant reduction in 30-day mortality was found in the wards (p = 0.205). In the emergency department, mortality remained stable in both phases of the study. In the ICU, we also observed a reduction, but without a statistically significant difference.

Figure 2

As we observed differences in the patients’ clinical characteristics, such as the frequency of neutropenia and acute leukemia, between the two phases of the study, we included the analysis of variables associated with 30-day and in-hospital mortality. In univariate analysis, SOFA (p=0.001), sepsis bundle department (wards and ICU) (p<0.001), non-community-acquired sepsis (p=0.001), neutropenia at sepsis diagnosis (p=0.011), abdominal source (p=0.036), septic shock (p<0.001), and time between first organ dysfunction and recognition of sepsis (p=0.014) were factors associated to 30-day mortality. Phase 2 and HSCT had higher frequency among survivors. As shown in Table 3, we observed lower sepsis bundle compliance rate among patients who progressed to death within 30 days (64.3%) compared to those who were alive (81.9%), with p<0.001. Age was associated with in-hospital mortality (p<0.001). Other factors associated with in-hospital mortality were similar to those found for 30-day mortality.

Table 3

Table 4. shows the variables associated to 30-day and in-hospital mortality in multivariate analysis. We observed that SOFA, sepsis bundle on wards or ICU, and abdominal infection were associated with 30-day and in-hospital mortality. Age was also associated with in-hospital mortality. Phase 2 and HSCT had OR lower than 1. Compliance with the sepsis bundle tended to be associated with lower in-hospital mortality (p=0.061).

Table 4

This study evaluated the impact of a multimodal educational program on sepsis bundle compliance and 30-day and in-hospital mortality in patients with hematologic malignancies, including HSCT recipients. We found an increase in sepsis bundle compliance in the 2nd phase of the study compared to the 1st phase. We also observed a reduction in the in-hospital mortality rate.

Several studies have evaluated the multimodal strategy in different scenarios, such as in sepsis, as well as the outcome of interventions in this population. Damiani E et al. (2015) conducted a systematic review in adult patients with sepsis, severe sepsis or septic shock that evaluated the impact of performance improvement program on compliance with guidelines based on Surviving Sepsis Campaign guidelines and/or mortality [11]. Interventions consisted of educational programs, process changes, or both. Performance improvement programs were associated with increased compliance with the full 6-hour bundle (OR = 4.12, p < 0.001) and the full 24-hour bundle (OR = 2.57, p < 0.001) and with reduced mortality (OR = 0.66, p < 0.001). Another systematic review [12], which evaluated the implementation of multimodal strategies to improve medical staff compliance to infection treatment guidelines, including sepsis, also demonstrated increased compliance rates. We observed in our study increased bundle compliance in the 2nd and 3rd phases compared to the 1st phase, but with a statistically significant difference only in the last phase. Although we found a reduction in mortality in 30 days, we did not observe any statistical difference.

Performance improvement programs are associated with increased compliance to resuscitation and sepsis management bundles and reduced mortality in patients with sepsis, severe sepsis, or septic shock [13, 2, 14, 15]. One of the positive points of our work was the creation of the sepsis team and the actions of the medical, nursing, and physical therapy leaders, educational intervention through playful training and realistic simulations obtained positive results. The current evidence proves that the effective implantation of management bundles impacts patient outcomes.

Among the risk factors that were associated with 30-day mortality, detection of sepsis in the wards and ICU was associated with 4x higher mortality when compared to the Emergency Room. This shows the impact of in-hospital sepsis and possibly failure of empiric antimicrobial therapy on the mortality of these patients.

Rivers E et al (2001), reinforce these findings, performed a prospective, randomized study revealed a reduction in the mortality rate from 46.5–30.5% (P = 0.009) in patients with sepsis and septic shock who opened the sepsis bundle and received directed therapy in the emergency room [16].

In our study, organ dysfunction was also associated with increased 30-day mortality, most notably renal dysfunction requiring acute hemodialysis. According to Reilly JP et al (2016), acute renal dysfunction is a frequent complication in cancer patients and is associated with an unfavorable prognosis, especially when associated with other organ dysfunctions [17].

Another point to be highlighted was the abdominal infection related to 30-day mortality. A study reinforces these findings, that the site of abdominal infection is associated with ICU and hospital mortality rates of 42 and 56%, respectively [18].

The multimodal strategy had an impact on increasing the rate of compliance to the sepsis bundle. In-hospital mortality significantly reduced in the intervention phase of the study. With our study, we demonstrate that the multimodal strategy can be a tool associated with improved compliance to the sepsis protocol and with a reduction in in-hospital mortality, which is important for the management of patients with hematological malignancies and sepsis.

SD

Standard deviation

SOFA

Sequential Organ Failure Assessment

qSOFA

quick Sequential Organ Failure

SIRS

Systemic inflammatory response syndrome

HSCT

Hematopoietic Stem Cell Transplant

ED

Emergency Department

ICU

Intensive Care Unit

Acknowledgements

Not applicable.

Authors’ contributions

Luciane Luz e Silva and Diogo Boldim Ferreira helped design the study, conduct the study, analyze and interpret the data, and draft and critically revise the manuscript. Karina Prandi Melillo and Roselene Mesquita Augusto Passos helped to conduct the study. Otávio Monteiro Becker Junior and Eduardo Alexandrino de Medeiros helped design the study, conduct the study, analyze and interpret the data, draft and critically revise the manuscript. All authors read and approved the final version of the manuscript.

Funding

No additional funding was received to complete this project.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from

the corresponding author on reasonable request.

Ethics approval and consent to participate

This project constituted an observation of standard care delivery with no randomization and thus met the definition of a service evaluation under the National Health Service Health research authority guidelines. As such, ethical approval was not required, and because all outcome measures are collected as part of routine care, the need for consent was waived. The study was approved by the Research Ethics Committee of the Federal University of São Paulo / Escola Paulista de Medicina, under the Certificate of Presentation for Ethical Appreciation (CAAE) number 15174719.9.0000.5505 and Ethics Committee of Hospital de Transplantes Euryclides de Jesus Zerbini, under the Certificate of Presentation for Ethical Appreciation (CAAE) number 15174719.9.3001.0091.

Consent for publication

Not applicable.

Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the infectious diseases society of America. Clin Infect Dis 2011;52(4):e56-93.
Machado FR, Cavalcanti AB, Bozza FA, Ferreira EM, Angotti Carrara FS, Sousa JL, Caixeta N, Salomao R, Angus DC, Pontes Azevedo LC; SPREAD Investigators; Latin American Sepsis Institute Network. The epidemiology of sepsis in Brazilian intensive care units (the Sepsis PREvalence Assessment Database, SPREAD): an observational study. Lancet Infect Dis. 2017 Nov;17(11):1180–1189.
Ferrer R, Artigas A, Suarez D, et al.: Effectiveness of treatments for severe sepsis: A prospective, multicenter, observational study. Am J Respir Crit Care Med 2009; 180:861–866.
Boucher JE, Carpenter D. Sepsis: Symptoms, Assessment, Diagnosis, and the Hour-1 Bundle in Patients With Cancer. Clin J Oncol Nurs. 2020 Feb 1;24(1):99–102.
Legrand M, Max A, Peigne V, Mariotte E, Canet E, Debrumetz A, Lemiale V, Seguin A, Darmon M, Schlemmer B, Azoulay E. Survival in neutropenic patients with severe sepsis or septic shock. Crit Care Med. 2012;40(1):43–9.
Ferrer R, Artigas A, Levy MM, Blanco J, González-Díaz G, Garnacho-Montero J, Ibáñez J, Palencia E, Quintana M, de la Torre-Prados MV; Edusepsis Study Group. Improvement in process of care and outcome after a multicenter severe sepsis educational program in Spain. JAMA. 2008 May 21;299(19):2294 – 303.
Evans L, Rhodes A, Alhazzani W, Antonelli M, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021 Nov;47(11):1181–1247.
CDC/NHSN Surveillance Definitions for Specific Types of Infections. January 2020.
R Core Team (2020) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/.
HOSMER, David W.; LEMESHOW (2000), Stanley. Applied logistic regression. 2ª ed. John Wiley & Sons.
Damiani E, Donati A, Serafini G, Rinaldi L, Adrario E, Pelaia P, Busani S, Girardis M. Effect of performance improvement programs on compliance with sepsis bundles and mortality: a systematic review and meta-analysis of observational studies. PLoS One. 2015 May 6;10(5):e0125827
McKenzie KE, Mayorga ME, Miller KE, Singh N, Arnold RC, Romero-Brufau S. Notice to comply: A systematic review of clinician compliance with guidelines surrounding acute hospital-based infection management. Am J Infect Control. 2020 Aug;48(8):940–947.
Hampshire PA, Welch CA, McCrossan LA, Francis K, Harrison DA. Admission factors associated with hospital mortality in patients with haematological malignancy admitted to UK adult, general critical care units: a secondary analysis of the ICNARC Case Mix Programme Database. Crit Care. 2009;13(4):R137.
Kuan WS, Mahadevan M, Tan JH, Guo J, Ibrahim I. Feasibility of introduction and implementation of the Surviving Sepsis Campaign bundle in a Singapore emergency department. Eur J Emerg Med. 2013;20(5):344–9.
Na S, Kuan WS, Mahadevan M, Li CH, Shrikhande P, Ray S, Batech M, Nguyen HB; ATLAS Investigators. Implementation of early goal-directed therapy and the surviving sepsis campaign resuscitation bundle in Asia. Int J Qual Health Care. 2012;24(5):452–62.
Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M. Early Goal- directed therapy in the treatment of severe sepsis and septic shock New England J Med, Vol. 345, No. 19 · November 8, 2001.
Reilly JP, Anderson BJ, Hudock KM, Dunn TG, Kazi A, Tommasini A, Charles D, Shashaty MG, Mikkelsen ME, Christie JD, Meyer NJ. Neutropenic sepsis is associated with distinct clinical and biological characteristics: a cohort study of severe sepsis. Crit Care. 2016;20(1):222.
Torres VB, Azevedo LC, Silva UV, et al. Sepsis-Associated Outcomes in Critically Ill Patients with Malignancies. Ann Am Thorac Soc. 2015;12(8):1185–1192.

Table 1. General patient data and comparison between study periods, January/2017 to August/2020.
Variables	Total	1st phase	2nd phase	p-valoue ⁽¹⁾
Variables	(n = 329)	(n = 101)	(n = 228)	p-valoue ⁽¹⁾
Demographic data
Female gender	165 (50.2%)	49 (48.5%)	116 (50.9%)	0.693
Age, mean in years ± SD	58.6 ± 15.2	57.7 ± 16.7	58.9 ± 14.5	0.523
SOFA, mean ± SD	5.1 ± 3.3	5 ± 3.3	5.1 ± 3.3	0.736
Sepsis diagnosis
Sepsis	241 (73.3%)	69 (68.3%)	172 (75.4%)	0.178
Septic shock	88 (26.7%)	32 (31.7%)	56 (24.6%)
Sepsis bundle department
ED	146 (44.4%)	42 (41.6%)	104 (45.6%)	0.497
Ward	164 (49.8%)	53 (52.5%)	111 (48.7%)	0.526
ICU	19 (5.8%)	6 (5.9%)	13 (5.7%)	0.932
Baseline Diagnosis
Lymphomas	80 (24.3%)	19 (18.8%)	61 (26.8%)	0.121
Multiple myeloma	65 (19.8%)	17 (16.8%)	48 (21.1%)	0.375
Chronic leukemias	49 (14.9%)	16 (15.8%)	33 (14.5%)	0.748
Acute leukemias	53 (16.1%)	24 (23.8%)	29 (12.7%)	0.012
Myelodysplastic syndrome	17 (5.2%)	6 (5.9%)	11 (4.8%)	0.673
Others	65 (19.8%)	19 (18.8%)	46 (20.2%)	0.774
HSCT	39 (11.9%)	9 (8.9%)	30 (13.2%)	0.272
Autologous	28 (71.8%)	4 (44.4%)	24 (80%)	0.085
Allogeneic	11 (28.2%)	5 (55.6%)	6 (20%)
Neutropenia	91 (27.7%)	36 (35.6%)	55 (24.1%)	0.031
Chemotherapy	189 (57.4%)	49 (48.5%)	140 (61.4%)	0.029
Suspected site of infection
Respiratory	171 (52.0%)	54 (53.5%)	117 (51.3%)	0.719
Abdomen	58 (17.6%)	13 (12.9%)	45 (19.7%)	0.132
Unknown	40 (12.2%)	16 (15.8%)	24 (10.5%)	0.174
Bloodstream	25 (7.6%)	5 (5%)	20 (8.8%)	0.228
Urinary tract	16 (4.9%)	5 (5%)	11 (4.8%)	0.259
Skin/soft tissue	8 (2.4%)	1 (1%)	7 (3.1%)	0.961
Others	11 (3.3%)	7 (6.9%)	4 (1.8%)	0.016
Community-acquired sepsis	161 (48.9%)	47 (46.5%)	114 (50%)	0.562
Mechanical ventilation	76 (23.1%)	29 (28.7%)	47 (20.6%)	0.108
Acute dialysis	26 (7.9%)	10 (10.3%)	16 (7.3%)	0.377
Outcomes
ICU admission	160 (50.6%)	54 (53.5%)	106 (49.3%)	0.490
ICU length of stay, mean in days ± SD	11.0 ± 17.1	16.3 ± 24	8.6 12.1	0.028
Hospital length of stay, mean in days ± SD	17.5 ± 18.2	20.5 ± 23.7	16.2 ± 15	0.095
30-day mortality	126 (38.3%)	43 (42.6%)	83 (36.4%)	0.288
In-hospital mortality	139 (42.2%)	52 (51.5%)	87 (38.2%)	0.024
Legend: SD: Standard deviation; SOFA: Sequential Organ Failure Assessment; qSOFA: quick Sequential Organ Failure; SIRS: Systemic inflammatory response syndrome; HSCT: Hematopoietic Stem Cell Transplant

⁽¹⁾ Fisher's tests were performed for qualitative variables with 2 categories, Wilcoxon-Mann-Whitney's test for numerical variables, and Chi-square tests for the remaining variables.

Table 2. Sepsis bundle compliance and comparison between study phases, January/2017 to August/2020.
Variables	Total (n = 329)	1st phase (n = 101)	2nd phase (n = 228)	p-valoue ⁽¹⁾
Variables	Total (n = 329)	1st phase (n = 101)	2nd phase (n = 228)	p-valoue ⁽¹⁾
Time between first organ dysfunction and recognition of sepsis, mean in minutes ± SD	114.2 ± 228.9	160.7 ± 258.9	93.6 ± 212.3	0.012
(a) Lactate monitoring	303 (92.1%)	89 (88.1%)	214 (93.9%)	0,075
(b) Antimicrobial therapy within 1h	305 (92.7%)	88 (87.1%)	217 (95.2%)	0,010
(c) Blood cultures drawn prior to antibiotics	300 (91.2%)	84 (83.2%)	216 (94.7%)	<0.001
(d) Resuscitation	223 (96.1%)	70 (94.6%)	153 (96.8%)	0,410
3-hour bundle adherence	270 (82.1%)	72 (71.3%)	198 (86.8%)	0,001
(e) Lactate Clearance	52 (80.0%)	27 (81.8%)	25 (78.1%)	0,710
(f) Vasopressor within 6h	77/88 (87.5%)	25 (92.6%)	52 (85.2%)	0,337
(g) Volume and tissue perfusion reassessment within 6h	105 (83.3%)	45 (88.2%)	60 (80%)	0,223
Sepsis bundle compliance	247 (75.1%)	69 (68.3%)	178 (78.1%)	0,059
Empiric antibiotic therapy guideline	309 (93.9%)	93 (92.1%)	216 (94.7%)	0,352
⁽¹⁾Fisher's tests were performed for qualitative variables with 2 categories, Wilcoxon-Mann-Whitney's test for numerical variables, and Chi-square tests for the remaining variables.

Table 3: Univariate analysis of variables possibly associated with 30-day and in-hospital mortality.
Variables	30-day mortality		p-value ⁽¹⁾	In-hospital mortality		p-value ⁽¹⁾
Variables	Death (n=126)	Live (n=203)	p-value ⁽¹⁾	Death (n=126)	Live (n=203)	p-value ⁽¹⁾
Demographic data
Female gender	58 (46%)	107 (52.7%)	0.239	63 (45.3%)	102 (53.7%)	0.134
Age, mean in years ± SD	59.9 ± 14.1	57.7 ± 15.9	0.218	61.5 ± 13.1	59.4 ± 16.3	<0.001
SOFA, mean ± SD	6.7 ± 3.5	4.1 ± 2.7	<0.001	6.5 ± 3.5	4 ± 2.7	0.004
Study phase
Phase 1	43 (34.1%)	58 (28.6%)	0.288	52 (37.4%)	49 (25.8%)	0.024
Phase 2	83 (65.9%)	145 (71.4%)		87 (62.6%)	141 (74.2%)
Sepsis bundle department
Emergency Room	33 (26.2%)	113 (55.7%)	<0.001	41 (29.5%)	105 (55.3%)	<0.001
Wards	78 (61.9%)	86 (42.4%)	0.001	83 (59.7%)	81 (42.6%)	0.002
ICU	15 (11.9%)	4 (2%)	<0.001	15 (10.8%)	4 (2.1%)	0.034
Baseline Diagnosis
Acute leukemias	25 (19.8%)	28 (13.8%)	0.147	26 (18.7%)	27 (14.2%)	0.105
Multiple Myeloma	25 (19.8%)	40 (19.7%)	0.976	29 (20.9%)	36 (18.9%)	0.666
Lymphomas	29 (23%)	51 (25.1%)	0.665	33 (23.7%)	47 (24.7%)	0.835
Myelodysplastic syndrome	4 (3.2%)	13 (6.4%)	0.198	5 (3.6%)	12 (6.3%)	0.271
Chronic leukemias	18 (14.3%)	31 (15.3%)	0.807	21 (15.1%)	28 (14.7%)	0.926
Other	25 (19.8%)	40 (19.7%)	0.976	25 (18%)	40 (21.1%)	0.490
HSCT	7 (5.6%)	32 (15.8%)	0.005	9 (6.5%)	30 (15.8%)	0.010
Autologous	4 (57.1%)	24 (75%)	0.379	5 (55.6%)	23 (76.7%)	0.238
Allogeneic	3 (42.9%)	8 (25%)		4 (44.4%)	7 (23.3%)
Sepsis Diagnosis
Septic shock	53 (42.1%)	35 (17.2%)	<0.001	58 (41.7%)	30 (15.8%)	<0.001
Sepsis	73 (57.9%)	168 (82.8%)		81 (58.3%)	160 (84.2%)
Neutropenia	45 (35.7%)	46 (22.7%)	0.010	47 (33.8%)	44 (23.2%)	0.033
Chemotherapy	77 (61.1%)	112 (55.2%)	0.290	82 (59%)	107 (56.3%)	0.628
Infection sources
Bloodstream	7 (5.6%)	18 (8.9%)	0.271	7 (5%)	18 (9.5%)	0.133
Respiratory	67 (53.2%)	104 (51.2%)	0.732	74 (53.2%)	97 (51.1%)	0.695
Urinary tract	5 (4%)	11 (5.4%)	0.552	5 (3.6%)	11 (5.8%)	0.361
Skin/soft tissue	0 (0%)	8 (3.9%)	0.024	0 (0%)	8 (4.2%)	0.014
Abdomen	32 (25.4%)	26 (12.8%)	0.004	36 (25.9%)	22 (11.6%)	0.001
Unknown	12 (9.5%)	28 (13.8%)	0.249	13 (9.4%)	27 (14.2%)	0.183
Other infections	3 (2.4%)	8 (3.9%)	0.444	4 (2.9%)	7 (3.7%)	0.688
Community-acquired sepsis	47 (37.3%)	114 (56.2%)	<0.001	53 (38,1%)	108 (56,8%)	<0.001
Time between first organ dysfunction and recognition of sepsis, mean in minutes ± SD	156.6 ± 268.5	87.8 ± 197.4	0.014	145.3 ± 258.5	91.4 ± 203.1	0.021
Sepsis bundle compliance	81 (64.3%)	166 (81.8%)	<0.001	90 (64.7%)	157 (82.6%)	<0.001
Empiric antibiotic therapy guideline	116 (92.1%)	193 (95.1%)	0.267	129 (92.8%)	180 (94.7%)	0.469
Legend: SD: Standard deviation; SOFA: Sequential Organ Failure Assessment; SIRS: Systemic inflammatory response syndrome; HSCT: Hematopoietic Stem Cell Transplant; ICU: Intensive Care Unit.

⁽¹⁾ Fisher's tests were performed for qualitative variables with 2 categories, Wilcoxon-Mann-Whitney's test for numerical variables, and Chi-square tests were performed for the remaining variables.

Table 4. Multivariate analysis of variables associated with 30-day and in-hospital mortality.
Variables	30-day mortality			In-hospital mortality
Variables	OR	95% C.I.	p-value	OR	95% C.I.	p-value
Age	-	-	-	1.028	1.009-1.047	0.004
Phase 2	0.750	0.422-1.333	0.327	0.447	0.252-0.794	0.006
Wards (Sepsis bundle department)	4.237	2.404-7.469	<0.001	3.806	2.170-6.676	<0.001
ICU (Sepsis bundle department)	10.624	3.008-37.518	<0.001	7.989	2.307-27.667	0.001
Abdominal	2.229	1.124-4.422	0.022	3.039	1.524-6.059	0.002
HSCT	0.217	0.085-0.551	0.001	0.309	0.129-0.738	0.008
Sepsis bundle compliance	0.636	0.345-1.170	0.146	0.558	0.303-1.028	0.061
SOFA	1.297	1.186-1.418	<0.001	1.275	1.166-1.395	<0.001
Neutropenia	1.425	0.786-2.583	0.243	1.247	0.687-2.263	0.468
Community-acquired sepsis	0.688	0.337-1.403	0.304	0.973	0.474-1.995	0.939
Acute leukemias	1.122	0.526-2.393	0.766	1.034	0.481-2.220	0.932
Time between first organ dysfunction and recognition of sepsis	1	0.999-1.001	0.646	1	0.999-1.001	0.755
Sepsis	0.897	0.442-1.817	0.681	0.671	0.331-1.360	0.268
Legend: SD: Standard Deviation; SOFA: Sequential Organ Failure Assessment; BMT: Bone Marrow Transplant; ICU: Intensive Care Unit.

No competing interests reported.

Impact of multimodal strategy on sepsis bundle compliance and in- hospital mortality in patients with hematologic malignancies

Status:

Version 1

Abstract

Figures

Introduction

Methods

Intervention

Outcomes

Definitions

Statistical analysis

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Version 1