Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Taeyoung Kong; Hye Sun Lee; So Young Jeon; Je Sung You; Jong Wook Lee; Hyun Soo Chung; Sung Phil Chung

doi:10.1371/journal.pone.0258619

. 2021 Oct 15;16(10):e0258619. doi: 10.1371/journal.pone.0258619

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Taeyoung Kong ¹, Hye Sun Lee ², So Young Jeon ², Je Sung You ^1,^*, Jong Wook Lee ³, Hyun Soo Chung ¹, Sung Phil Chung ¹

Editor: Wisit Cheungpasitporn⁴

PMCID: PMC8519472 PMID: 34653202

Abstract

Background

Postpartum hemorrhage (PPH) constitutes a major risk for maternal mortality and morbidity. Unfortunately, the severity of PPH can be underestimated because it is difficult to accurately measure blood loss by visual estimation. The delta neutrophil index (DNI), which reflects circulating immature granulocytes, is automatically calculated in hematological analyzers. We evaluated the significance of the DNI in predicting hemorrhage severity based on the requirement for massive transfusion (MT) in patients with PPH.

Methods

We retrospectively analyzed data from a prospective registry to evaluate the association between the DNI and MT. Moreover, we assessed the predictive ability of the combination of DNI and shock index (SI) for the requirement for MT. MT was defined as a transfusion of ≥10 units of red blood cells within 24 h of PPH. In total, 278 patients were enrolled in this study and 60 required MT.

Results

Multivariable logistic regression revealed that the DNI and SI were independent predictors of MT. The optimal cut-off values of ≥3.3% and ≥1.0 for the DNI and SI, respectively, were significantly associated with an increased risk of MT (DNI: positive likelihood ratio [PLR] 3.54, 95% confidence interval [CI] 2.5–5.1 and negative likelihood ratio [NLR] 0.48, 95% CI 0.4–0.7; SI: PLR 3.21, 95% CI 2.4–4.2 and NLR 0.31, 95% CI 0.19–0.49). The optimal cut-off point for predicted probability was calculated for combining the DNI value and SI value with the equation derived from logistic regression analysis. Compared with DNI or SI alone, the combination of DNI and SI significantly improved the specificity, accuracy, and positive likelihood ratio of the MT risk.

Conclusion

The DNI and SI can be routinely and easily measured in the ED without additional costs or time and can therefore, be considered suitable parameters for the early risk stratification of patients with primary PPH.

Introduction

Postpartum hemorrhage (PPH) occurs in approximately 3–5% of obstetric patients [1]. Although it is uncommon, in approximately 20% of cases, PPH develops in patients without risk factors of bleeding [1]. PPH constitutes a major risk for maternal mortality and morbidity in both developed and developing countries and causes 12% of maternal mortality in the United States [1,2]. Generally, patients with a healthy pregnancy can tolerate blood loss of 500–1,000 mL with no symptoms or signs of hypovolemia [1,3]. PPH is defined as a blood loss of ≥ 500 mL, and severe PPH is defined as a blood loss of ≥ 1,000 mL. According to the American College of Obstetricians and Gynecologists, early PPH is defined as a total blood loss of or blood loss with symptoms and signs of hypovolemia of at least 1,000 mL within 24 h after intrapartum loss or delivery of fetus [1,4]. Primary PPH may develop before placenta delivery and up to 24 h after fetus delivery [1]. Prompt diagnosis and immediate resuscitation using a multidisciplinary team are required for the proper management of PPH [1,2]. Unfortunately, its severity can be underestimated because it is difficult to accurately measure blood loss by visual estimation, and there are no generally accepted cut-off limits for estimated blood loss [5]. Inaccurate estimation of blood loss can potentially lead to misleading clinical decisions [6,7]. Thus, objective parameters are needed for predicting hemorrhage severity and the requirement for massive transfusion in PPH patients [7]. The decrease in hematocrit is poorly correlated with blood loss severity and is clinically unavailable in emergency situations [5].

As traditional vital signs alone do not provide a reliable measurement of blood loss severity in the early stage after acute blood loss, some studies have demonstrated that the initial shock index (SI), defined as the ratio of the heart rate (HR) to the systolic blood pressure (SBP), was an independent predictor of the requirement for massive transfusion in patients with PPH [8,9]. An SI of >1.0 showed a positive likelihood ratio (PLR) of 2.8 and a negative likelihood ratio (NLR) of 0.5 for the requirement for massive transfusion in patients with PPH, indicating that the SI is of little help to clinicians for predicting the need for transfusion [7]. In general, likelihood ratios are only useful if their values are >5 or <0.2 [10]. Therefore, additional biomarkers or indicators are needed to better predict the requirement for massive transfusion in patients with PPH in the emergency department (ED). The DNI, which reflects the proportion of circulating immature granulocytes, is automatically calculated as the difference between the leukocyte differentials measured in the myeloperoxidase channel and those detected in the lobularity channel [11,12]. Many recent studies have founded that the DNI was significantly associated with the severity of diseases related to sterile inflammation, such as out-of-hospital cardiac arrest, major traumatic injury, pulmonary embolism, acute myocardial infarction, and upper gastrointestinal hemorrhage [12–16]. As an inflammatory response to hypoperfusion and reperfusion owing to blood loss and tissue damage can develop into systemic inflammatory response syndrome, the DNI is a promising candidate for a biomarker for predicting the requirement of massive transfusion in patients with primary PPH in the ED [12,13].

To the best of our knowledge, the usefulness of the DNI to stratify the severity of blood loss in patients with primary PPH in emergency settings has not been investigated. The aims of the present study were as follows: 1) to evaluate the clinical utility of the DNI as an indicator of disease severity based on the requirement for massive transfusion in patients with primary PPH admitted to the ED and 2) to investigate whether the combination of the initial DNI and SI, the latter being a hemodynamic indicator that can be easily obtained in the ED, can show improved predictive ability for the requirement for massive transfusion.

Methods

Study design and patients

This observational cohort study was retrospectively conducted based on a prospective registry of patients who presented with PPH in the ED of Severance Hospital between January 2011 and December 2019. The hospital is affiliated with Yonsei University College of Medicine and is a 2,500-bed tertiary-level referral hospital that serves approximately 95,000 patients per year. The study was approved by the institutional review board (IRB) of Yonsei University Health System (Seoul, Republic of Korea) (approval no.: 3-2017-0302), and the requirement for written informed consent was waived because of the retrospective nature of the study.

Since January 2009, the Severance Protocol to save postpartum blEeding through Expeditious care Delivery (SPEED) has been implemented as a critical pathway (CP) program based on computerized physician order entry (CPOE) at our institution as a part of a quality improvement initiative [2]. To provide prompt treatment and rapid transfusion to patients with primary PPH, we designed a standard protocol with an integrated alarm system using CPOE to reduce unnecessary delays in the hospital by implementing rapid decision-making by specialists. Primary PPH was defined as a hemorrhage requiring fluid resuscitation or transfusion within the first 24 h of delivery [7]. When a patient with PPH was approved for transfer from another hospital or a private clinic to our ED, emergency physicians and obstetricians were standing by in the ED to prepare for treatment, and the blood bank prepared universal O-type blood for transfusion. According to the predetermined protocol, an anesthesiologist and radiologist were also notified to prepare for emergency surgery or radiologic intervention using the CPOE and an automatic short message server [2]. Our institution schedules specialized obstetrics and interventional radiologists to be on call 24 h/day, 7 d/week, to treat patients who arrive at the ED requiring emergency surgical and radiological interventions [2]. Data on the patients’ clinical characteristics, laboratory results, severity of the clinical condition, and treatment process were prospectively obtained from the SPEED CP registry using a predetermined protocol.

This study analyzed consecutive patients with primary PPH who were prospectively integrated into the SPEED CP program from among those admitted between January 1, 2011 and December 31, 2019. Hence, patients who were referred to our ED for the evaluation and management of PPH from other hospitals or private obstetric clinics within the first 24 h of delivery were also included in this study. We excluded patients who met any of the following exclusion criteria: cardiac arrest before ED admission, refusal to receive a transfusion, and a diagnosis of hemolysis, elevated liver enzymes, and low platelets syndrome. Fig 1 shows the enrollment, exclusion, and clinical outcome data of patients with PPH.

Data collection

After obtaining the IRB approval, we accessed the SPEED registry database and collected data in January 2020. Data on the baseline and clinical characteristics of all patients; including age, parity, type of delivery, initial mental status, initial vital signs, initial laboratory findings, and amount of blood transfusion; and the clinical outcomes; including emergency surgery, embolization, hysterectomy, length of hospital stay, and in-hospital death; were retrieved from the prospective SPEED registry. Mental status was assessed at triage on ED admission using the Alert/Verbal/Painful/Unresponsive scale. Vital signs, including HR, body temperature, SBP, and diastolic BP, were also assessed at triage on ED admission. During the study period, emergency physicians or obstetricians treating the patient decided on the necessity, type, and amount of blood transfusion because there were no transfusion guidelines for patients with primary PPH and massive transfusion in our institution. Venous blood was collected in ethylenediaminetetraacetic acid-containing vacutainers within 10 min of ED admission for the measurement of the DNI using an automated blood cell analyzer (ADVIA 2120; Siemens, Forchheim, Germany) used for CBC analysis. We also performed other laboratory tests, including tests for prothrombin time, total CO₂, blood urea nitrogen, creatinine levels, albumin levels, and total bilirubin levels, at the time of ED admission.

Patients’ characteristics

In total, 278 patients with primary PPH were enrolled in this study. Of those, 60 patients (21.6%) required massive transfusion (Fig 1). Table 1 shows the baseline and clinical characteristics of patients classified according to the requirement for massive transfusion. There was no difference in parity between the two groups. However, there were significant differences in age, body temperature, mental status on ED admission, and type of delivery between the two groups. Patients who required a massive transfusion had a significantly lower SBP (94 ± 26 vs. 115 ± 20 mmHg, p<0.001) and diastolic BP (59 ± 18 vs. 69 ± 13 mmHg, p<0.001) and a higher HR (114 ± 23 vs 96 ± 20 bpm, p<0.001) on ED admission than those who did not require a massive transfusion. The initial SI on ED admission was significantly higher in patients who required a massive transfusion than in those who did not require a massive transfusion (1.30 ± 0.48 vs 0.86 ± 0.24, p<0.001). Patients who received a massive transfusion had a higher DNI than those who did not (5.21 ± 4.95% vs 1.86 ± 1.76%, p<0.001). In addition, they had lower hemoglobin, platelet, total CO₂, and albumin levels. Considering therapeutic management, patients who received a massive transfusion required more units of blood components and required emergency surgery and hysterectomy more often than those who did not receive a massive transfusion. In addition, they had a longer ICU admission (1.7 ± 1.6 vs. 0.3 ± 0.7 days, p<0.001). However, in-hospital mortality was not significantly different between the two groups; only one case of death was observed in our study, which occurred in a patient who required a massive transfusion.

Table 1. Comparison of clinical characteristics and clinical outcomes according to the requirement for massive transfusion in patients with primary postpartum hemorrhage.

Variables	Total	Non-MT	MT	P
Variables	N = 278	N = 218 (78.4%)	N = 60 (21.6%)	P
Age (years)	32.70±3.95	32.37±3.93	34.03±3.78	0.004^*
Parity [n (%)]				0.46
Primipara	169(60.79)	135(61.93)	34(56.67)
Multipara	109(39.21)	83(38.07)	26(43.33)
Type of delivery [n (%)]				0.026^*
Vaginal delivery	182(65.47)	150(68.81)	32(53.33)
Caesarean section	96(34.53)	68(31.19)	28(46.67)
Initial mental status [n (%)]				<0.001^*
Alert	271(97.48)	217(99.54)	54(90.00)
Non-alert	7(2.52)	1(0.46)	6(10.00)
Initial vital sign
Shock index	0.95±0.35	0.86±0.24	1.30±0.45	<0.001^*
Shock index > 1 [n (%)]	94 (33.81)	48 (22.02)	46 (76.67)	<0.001^*
SBP (mmHg)	111±23	115±20	94±26	<0.001^*
DBP (mmHg)	67±15	69±13	59±18	<0.001^*
Heart rate (bpm)	100±22	96±20	114±23	<0.001^*
Body temperature (°C)	36.8±0.7	36.9±0.7	36.5±0.8	0.002^*
Initial laboratory data
WBC count (10^3/μL)	16.94±6.72	16.94±6.68	16.92±6.94	0.984
Hemoglobin (g/dL)	9.67±2.03	9.91±1.94	8.81±2.13	<0.001^*
Platelet count (10^3/μL)	180.54±84.23	194.94±84.03	128.23±61.56	<0.001^*
Albumin (g/dL)	2.37±0.62	2.50±0.57	1.98±0.60	<0.001^*
Total CO2 (mmol/L)	17.74±3.08	18.17±2.80	16.19±3.52	<0.001^*
Arterial lactate (mmol/L)	2.71±2.39	1.94±0.96	3.80±3.24	<0.001^*
Delta neutrophil index	2.58±3.09	1.86±1.76	5.21±4.95	<0.001^*
Blood transfusion
Transfusion before ED arrival [n (%)]	130(46.76)	104(43.33)	26(68.42)	0.004^*
pRBCs before ED arrival (unit)	1.66±2.90	1.27±2.10	4.11±5.23	0.002^*
Total pRBCs (unit)	6.32±7.06	3.39±2.46	16.98±8.04	<0.001^*
Total fresh frozen plasma (unit)	2.04±3.81	0.68±1.42	6.95±5.42	<0.001^*
Total platelet concentrate (unit)	2.49±6.52	0.62±2.21	9.27±11.05	<0.001^*
Clinical outcome [n (%)]
Embolization	199(71.58)	151(69.27)	48(80.00)	0.103
Hysterectomy	15(5.40)	1(0.46)	14(23.33)	<0.001^*
Emergent surgery	42(15.11)	9(4.13)	33(55.00)	<0.001^*
In-hospital death	1(0.36)	0(0.00)	1(1.67)	0.216

Open in a new tab

*P<0.05. MT, massive transfusion; SBP, systolic blood pressure; DBP, diastolic blood pressure; WBC, white blood cell; ED, emergency department; pRBCs, packed red blood cells.

DNI measurement

The automated blood cell analyzer could determine the CBC count, comprising the DNI, white blood cell (WBC) count, hemoglobin level, and platelet count [12,17]. The specific analyzers used to determine the DNI were optical systems based on a cytochemical myeloperoxidase tungsten-halogen channel (which evaluates and differentiates eosinophils, lymphocytes, monocytes, neutrophils, and large unstained cells based on myeloperoxidase staining intensity and cell size) and a laser-diode channel (which calculates, counts, and classifies cell types based on lobularity/nuclear density and size) [12,17–19]. The DNI was then calculated by subtracting the fraction of mature polymorphonuclear neutrophils from the sum of myeloperoxidase-reactive cells, detecting circulating immature granulocytes as a leukocyte sub-fraction [12,17,18]. The DNI was obtained using the following formula: DNI = (neutrophil sub-fraction + eosinophil sub-fraction measured in the myeloperoxidase channel)—(polymorphonuclear sub-fraction measured in the nuclear lobularity channel) [11–13].

SI measurement

The following formula was used for calculating the SI: SI = HR/SBP. The SI on ED admission was calculated based on the vital signs on ED admission.

Clinical endpoints

The primary endpoint was the requirement for massive transfusion within 24 h of the development of PPH. Massive transfusion was defined as a transfusion of ≥ 10 units of packed red blood cells. To determine the amount of transfusion, we measured both the amount of transfused blood before arrival at the ED and the amount of transfused blood after arrival at the ED (in the ED and in the intensive care unit [ICU] or general ward).

Statistical analysis

The data are presented as mean (SD) for continuous variables and as absolute or relative frequencies for categorical variables. Continuous variables were compared using the two-sample t-test or Mann-Whitney U-test, and categorical variables were compared using the χ2 test or Fisher’s exact test. Univariate logistic regression analysis was performed to identify relationships between demographic characteristics and clinical data. A multivariate logistic regression analysis integrated the major covariates identified in the univariate analyses (variables with a p<0.05). A multivariate logistic regression analysis was also performed to assess promising independent factors for predicting the requirement for massive transfusion. We plotted receiver operating characteristic (ROC) curves and determined the areas under the curve (AUCs) to evaluate the ability of the DNI and SI to predict the requirement for massive transfusion. Youden’s method was used to verify the optimal cut-off of the DNI and SI for discriminating between the requirement and non-requirement for massive transfusion. These results are presented as odds ratios (ORs) and 95% confidential intervals (CIs). In addition, we computed the ROC curve to determine the predictive accuracy and identified the sensitivity, specificity, accuracy, PLR, and NLR for predicting the requirement for massive transfusion considering the DNI, SI, and the combination of the DNI and SI; we also calculated the optimal cut-off values. In addition, to stratify risk for the requirement for massive transfusion by combining the DNI value and SI value, we used the equation derived from logistic regression analysis. Based on this equation, we calculated the predicted probability for combining the DNI value and SI value. We determined optimal the cut-off point for the predicted probability. Based on this optimal cut-off point, we measured sensitivity, specificity, PLR, NLR, and accuracy. The generalized estimating equation (GEE) method was used to compare the significance of the DNI, SI, and the combination of the DNI and SI. Statistical analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA) and MedCalc Statistical Software version 16.4.3 (MedCalc Software bvba, Ostend, Belgium), and a p value < 0.05 was considered significant.

Results

DNI and SI as predictors of the requirement for massive transfusion

Multivariate logistic regression analysis was used to identify independent factors associated with the requirement for massive transfusion (Table 2). Based on the univariate analysis results, initial mental status, hemoglobin level, serum total CO₂ level, albumin concentration, SI, and DNI were included in the multivariate logistic regression analysis (S1 Table). The requirement for massive transfusion was not independently associated with the hemoglobin and serum total CO₂ levels on ED admission. The DNI value was independently associated with the requirement for massive transfusion (OR: 1.247, 95% CI: 1.072–1.451, p = 0.004).

Table 2. Factors associated with the requirement for massive transfusion in patients with primary postpartum hemorrhage in multivariate analysis.

Variable	Unadjusted OR (95% CI)	P	Adjusted OR^† (95% CI)	P
Hemoglobin (per 1g/dL)	0.759 (0.653–0.881)	<0.001^*	0.911 (0.745–1.114)	0.365
Total CO2 (per 1mmol/L)	0.802 (0.723–0.888)	<0.001^*	0.990 (0.865–1.132)	0.881
Albumin (per 1g/dL)	0.181 (0.097–0.339)	<0.001^*	0.372 (0.172–0.805)	0.012^*
Mental change (vs alert)	24.108 (2.843–204.450)	0.004^*	14.744 (1.319–164.838)	0.029^*
Delta neutrophil index	1.490 (1.305–1.701)	<0.001^*	1.247 (1.072–1.451)	0.004^*
Shock index	56.292 (17.269–183.498)	<0.001^*	13.730 (3.462–54.458)	<0.001^*

Open in a new tab

*P<0.05. OR, odds ratio; CI, confidence interval

†Multivariate model including all the variables listed in the table.

The SI was also independently associated with the requirement for massive transfusion (OR: 13.73, 95% CI: 3.46–54.46; p<0.001). In the ROC curve analysis of the requirement for massive transfusion, the AUC values of the DNI and SI were 0.74 (95% CI: 0.66–0.82; p<0.001) and 0.82 (95% CI: 0.76–0.89; p<0.001), respectively (Fig 2). Using Youden’s method, the optimal cut-off values for the DNI and SI on ED admission were found to be 3.3% and 1, respectively.

Fig 2 — DNI, delta neutrophil index; SI, shock index; AUC, area under the curve; CI, confidence interval.

An increase in the DNI of ≥3.3% was significantly associated with the requirement for massive transfusion, with a specificity of 83% (95% CI: 78–88), sensitivity of 60% (95% CI: 47–72), accuracy of 78% (95% CI: 73–83), PLR of 3.54 (95% CI; 2.47–5.06), and NLR of 0.48 (95% CI; 0.35–0.66) (Table 3). An increase in the SI of ≥ 1.0 was also significantly associated with the requirement for massive transfusion, with a specificity of 76% (95% CI: 70–82), sensitivity of 77% (95% CI: 66–87), accuracy of 76% (95% CI: 71–81), PLR of 3.21 (95% CI: 2.44–4.23), and NLR 0.31 (95% CI: 0.19–0.49).

Table 3. Prognostic value of the shock index, delta neutrophil index, and combined delta neutrophil index with shock index for predicting the requirement for a massive transfusion in patients with primary postpartum hemorrhage.

Variable	Sensitivity (95% CI)	Specificity (95% CI)	Accuracy (95% CI)	PLR	NLR
Variable	Sensitivity (95% CI)	Specificity (95% CI)	Accuracy (95% CI)	(95% CI)	(95% CI)
1> Delta neutrophil index ≥ 3.3	60	83	78.1	3.54	0.48
1> Delta neutrophil index ≥ 3.3	(47.6–72.4)	(78.1–88.0)	(73.2–82.9)	(2.47–5.06))	(0.35–0.66)
2> Shock index ≥ 1	76.7	76.2	76.3	3.21	0.31
2> Shock index ≥ 1	(66.0–87.4)	(70.5–81.8)	(71.3–81.3)	(2.44–4.23)	(0.19–0.49)
3> Predicted probability^§ ≥ 0.249	70	86.7	83.1	5.26	0.35
3> Predicted probability^§ ≥ 0.249	(58.4–81.6)	(82.2–91.2)	(78.7–87.5)	(3.61–7.67)	(0.23–0.51)
	Comparison of P value
1 vs 3	0.038^*	<0.001^*	<0.001^*	<0.001^*	0.853
2 vs 3	<0.001^*	<0.001^*	<0.001^*	<0.001^*	0.015^*

Open in a new tab

*P<0.05. PLR, positive likelihood ratio; NLR, negative likelihood ratio; CI, confidence interval.

§Equation of predicted probability = 1/(1+exp (5.348–0.2464*DNI-3.2691*SI).

Prognostic value of the combination of the DNI and SI for the requirement for massive transfusion

Considering the results for the SI and DNI on ED admission, we found that the SI had a lower specificity and the DNI had a lower sensitivity and a higher specificity for predicting the requirement for massive transfusion. To compensate for the weaknesses of both indicators, we measured the sensitivity, specificity, accuracy, PLR, and NLR of the combination of the DNI and SI. On combining a DNI value of ≥3.3 with an SI value of ≥1.0, the specificity and PLR increased to 95% (95% CI; 93–98) and 11.63 (95% CI: 6.07–22.27), respectively. However, the sensitivity and NLR were 53% (95% CI: 41–66) and 0.49 (95% CI: 0.37–0.64), respectively.

The predicted probability derived from the logistic equation was computed for each patient. When the predicted probability of the equation [Predicted probability = 1/(1+exp(5.348–0.2464*DNI-3.2691.*SI)] containing the combination of the DNI and SI was more than 0.249, the specificity was 87% (95% CI: 82–91), sensitivity was 70% (95% CI: 58–82), accuracy was 83% (95% CI: 79–88), PLR was 5.26 (95% CI: 3.6–7.7), and NLR was 0.35 (95% CI: 0.2–0.5) for MT. The combination of the DNI and SI significantly improved the specificity (DNI vs DNI + S, p<0.001; SI ≥1.1 vs DNI + SI, p<0.001), accuracy (DNI vs DNI + SI, p<0.001; SI ≥1.1 vs DNI + SI, p<0.001), and PLR (DNI vs DNI + S, p<0.001; SI ≥1.1 vs DNI + SI, p = 0.038; using the GEE method) for the requirement for massive transfusion when compared with either DNI or SI (Table 3).

Discussion

We demonstrated that the DNI on ED admission, which reflects the fraction of immature granulocytes, was independently associated with the requirement for massive transfusion in patients with primary PPH, with an AUC value of 0.74. Using the optimal cut-off value of 3.3% for the DNI on ED admission, the DNI was significantly associated with an increased risk of massive transfusion, with a PLR of 3.54 and an NLR of 0.48. To the best of our knowledge, this study is the first to identify the usefulness of the DNI on ED admission for predicting the requirement for massive transfusion in patients with primary PPH.

The optimal cut-off value for the DNI on ED admission was 3.3% in this study. Yune et al. found that a DNI >10.5% on day 1 significantly predicted a higher 30-day mortality rate after out-of-hospital cardiac arrest [16]. In a study on trauma patients admitted to an ICU, Kong et al. proposed that DNI values of >3.25% and >5.3% at 12 h post-admission were significantly associated with the development of multiple organ dysfunction syndrome and 30-day mortality, respectively [12]. Kong et al. also demonstrated that the optimal cut-off value of the DNI was 1% at admission (p<0.001) and 2.6% on day 1 (p<0.001) for predicting 30-day mortality in patients with upper gastrointestinal bleeding [13]. Taken together, these studies suggested that the DNI values over time reflect the severity of diseases related to systemic and sterile inflammation without infection [12].

The DNI is available in emergency settings because it can be rapidly and easily determined using an automated blood cell analyzer while measuring the CBC [11–13] The DNI value is correlated with the manual count of immature granulocytes [11]. However, automatic calculation of the DNI is a more prompt and accurate method than manual counting of immature granulocytes [12]. Several studies have shown that an increase in the DNI can predict the severity of infection and sterile inflammation without infection and that it is significantly associated with poor outcomes in specific disease conditions [12–14,17,18,20,21]. Emergency granulopoiesis is the most accepted theory for explaining why an elevated DNI value is associated with the use of massive transfusion after primary PPH [22,23]. Emergency granulopoiesis implies the expansion of the immature granulocytic population for providing hematopoietic precursors to sustain the increasing number of circulating neutrophils [12,22,23]. The number of circulating immature neutrophils can be rapidly increased to compensate for the destruction of mature neutrophils due to severe systemic inflammation and the loss of active neutrophils due to massive consumption related to sterile stimuli, such as chemical agents, trauma, bleeding, and ischemic insult [22,24,25]. Livingston et al. proposed that bone marrow failure can develop due to insufficient perfusion and ischemia during hemorrhagic shock [26,27]. This bone marrow dysfunction causes a transient failure of granulopoiesis [26,27]. Consequently, severe hemorrhage may contribute to the release of immature granulocytes into the circulation [24,28]. As pregnant women often have prenatal anemia, it is difficult to determine the occurrence of massive hemorrhage and evaluate the requirement for massive transfusion based on hemoglobin levels alone in patients with PPH [29]. Although lower hemoglobin or hematocrit levels have been widely and interchangeably used as indicators of severe bleeding, their diagnostic values for determining the severity of blood loss in the initial phase in patients with severe hemorrhage, such as trauma-related and GI bleeding, remain controversial [29]. In the present study, the hemoglobin level on ED admission was not associated with the requirement for massive transfusion.

In the present study, we found that using a combination of the DNI and SI significantly improved the specificity, accuracy, and PLR for the requirement for massive transfusion in patients with primary PPH when compared with either DNI or SI.

As PPH is significantly associated with a high risk of maternal death, it is crucial to assess and hemodynamically stabilize patients in a timely manner by careful monitoring [2]. In high-risk patients, prompt resuscitation and treatment should be initiated to decrease maternal mortality and morbidity [2]. Compared with the findings of Sohn et al., in our study, the SI had a relatively high sensitivity. This suggested its clinical implication: the initial SI value can be used as a screening tool for the initial risk stratification on ED admission, while the combination of the SI and DNI can be used as an additional prediction tool for the requirement for massive transfusion in patients with primary PPH.

This study had several limitations. First, although the data were obtained from the prospectively maintained SPEED CP registry using a standardized and predetermined protocol at our institution, they were obtained from a single, tertiary, academic hospital and were analyzed retrospectively. Furthermore, as the present study included patients with PPH who were transferred from other hospitals or private clinics to our ED, we had to rely on the medical records of the referring hospital or clinic to identify the amount of transfused blood before ED admission at our institution. Therefore, it was difficult to completely control for confounding factors, which increased the potential risk of selection bias. Second, we could not obtain accurate data related to the time of development of massive bleeding in the referring hospitals and clinics. For minimizing the effect of prehospital transfusions, we included patients with ED admission within 24 h of the onset of primary PPH. Third, a previous study demonstrated that the DNI value was an independent predictor of the development of multiple organ dysfunction syndrome and short-term mortality in trauma patients who were admitted to an ICU, but its predictive ability was not affected by the amount of transfusion [12]. As transfused blood products may contain WBCs, prehospital transfusions may have affected the DNI value. However, the WBCs in the blood products may get diluted in the blood of the whole body; thus, the effect on the DNI value may be trivial. Nevertheless, further prospective, randomized, multicenter trials are required to assess the prehospital time factor, the effect of prehospital transfusions, and the clinical applicability of our results.

Conclusions

We demonstrated that an increased DNI is an independent predictor of the requirement for massive transfusion in patients with primary PPH admitted to the ED. Combining the initial DNI with the initial SI can improve the predictive ability for the requirement for massive transfusion. The DNI and SI can be routinely and easily measured in the ED without additional costs or time and can therefore, be considered suitable parameters for the early risk stratification of patients with primary PPH. In terms of clinical implications, the initial SI value can be useful as a screening tool for risk stratification on ED admission, and the combination of the SI and DNI can be applied as a confirmative prediction tool for the requirement for massive transfusion in patients with primary PPH.

Supporting information

S1 Table. Univariate logistic regression analysis for clinical factors associated with the requirement for massive transfusion in patients with primary postpartum hemorrhage.

(DOCX)

Click here for additional data file.^{(18.7KB, docx)}

S2 Table. Original data that was used in the study.

(XLSX)

Click here for additional data file.^{(149.1KB, xlsx)}

Abbreviations

PPH: Postpartum hemorrhage
DNI: delta neutrophil index
SI: shock index
MT: massive transfusion
ORs: odds ratios
HR: heart rate
SBP: systolic blood pressure
ED: emergency department
CBC: complete blood count
SPEED: postpartum blEeding through Expeditious care Delivery
CP: critical pathway
CPOE: computerized physician order entry
WBC: white blood cell
ICU: intensive care unit
ROC: receiver operating characteristic
AUCs: areas under the curve
Cis: confidential intervals
GEE: generalized estimating equations

Data Availability

All relevant data are within the manuscript and its Supporting Information files.

Funding Statement

This study was supported by the Basic Science Research Program of the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT, in the form of a grant awarded to JSY (NRF- 2018R1C1B6006159), Yonsei University College of Medicine in the form of a faculty research grant for 2019 awarded to JSY (6-2019-0188), and Siemens Health Care in the form of research funding awarded to JSY (3-2015-0140), which did not exceed $10,000 per year. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Evensen A, Anderson JM, Fontaine P. Postpartum Hemorrhage: Prevention and Treatment. Am Fam Physician. 2017;95: 442–449 . [PubMed] [Google Scholar]
2.Cho HY, Na S, Kim MD, Park I, Kim HO, Kim YH, et al. Implementation of a multidisciplinary clinical pathway for the management of postpartum hemorrhage: a retrospective study. Int J Qual Health Care. 2015;27: 459–465. doi: 10.1093/intqhc/mzv068 . [DOI] [PubMed] [Google Scholar]
3.Magann EF, Evans S, Hutchinson M, Collins R, Lanneau G, Morrison JC. Postpartum hemorrhage after cesarean delivery: an analysis of risk factors. South Med J. 2005;98: 681–685. doi: 10.1097/01.SMJ.0000163309.53317.B8 . [DOI] [PubMed] [Google Scholar]
4.Girault A, Deneux-Tharaux C, Sentilhes L, Maillard F, Goffinet F. Undiagnosed abnormal postpartum blood loss: Incidence and risk factors. PLoS One. 2018;13: e0190845. doi: 10.1371/journal.pone.0190845 . [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Rath WH. Postpartum hemorrhage—update on problems of definitions and diagnosis. Acta Obstet Gynecol Scand. 2011;90: 421–428. doi: 10.1111/j.1600-0412.2011.01107.x . [DOI] [PubMed] [Google Scholar]
6.Sohn CH, Kim SR, Kim YJ, Seo DW, Ahn S, Lee YS, et al. Disseminated Intravascular Coagulation in Emergency Department Patients With Primary Postpartum Hemorrhage. Shock. 2017;48: 329–332. doi: 10.1097/SHK.0000000000000852 . [DOI] [PubMed] [Google Scholar]
7.Sohn CH, Kim YJ, Seo DW, Won HS, Shim JY, Lim KS, et al. Blood lactate concentration and shock index associated with massive transfusion in emergency department patients with primary postpartum haemorrhage. Br J Anaesth. 2018;121: 378–383. doi: 10.1016/j.bja.2018.04.039 . [DOI] [PubMed] [Google Scholar]
8.Sohn CH, Kim WY, Kim SR, Seo DW, Ryoo SM, Lee YS, et al. An increase in initial shock index is associated with the requirement for massive transfusion in emergency department patients with primary postpartum hemorrhage. Shock. 2013;40: 101–105. doi: 10.1097/SHK.0b013e31829b1778 . [DOI] [PubMed] [Google Scholar]
9.Borovac-Pinheiro A, Ribeiro FM, Morais SS, Pacagnella RC. Shock index and heart rate standard reference values in the immediate postpartum period: A cohort study. PLoS One. 2019;14: e0217907. doi: 10.1371/journal.pone.0217907 . [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Perera R, Heneghan C. Making sense of diagnostic test likelihood ratios. ACP J Club. 2007;146: A8–9 . [PubMed] [Google Scholar]
11.Nahm CH, Choi JW, Lee J. Delta neutrophil index in automated immature granulocyte counts for assessing disease severity of patients with sepsis. Ann Clin Lab Sci. 2008;38: 241–246 . [PubMed] [Google Scholar]
12.Kong T, Park YS, Lee HS, Kim S, Lee JW, You JS, et al. The delta neutrophil index predicts development of multiple organ dysfunction syndrome and 30-day mortality in trauma patients admitted to an intensive care unit: a retrospective analysis. Sci Rep. 2018;8: 17515. doi: 10.1038/s41598-018-35796-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Kong T, In S, Park YS, Lee HS, Lee JW, You JS, et al. Usefulness of the Delta Neutrophil Index to Predict 30-Day Mortality in Patients with Upper Gastrointestinal Bleeding. Shock. 2017;48: 427–435. doi: 10.1097/SHK.0000000000000878 . [DOI] [PubMed] [Google Scholar]
14.Kong T, Kim TH, Park YS, Chung SP, Lee HS, Hong JH, et al. Usefulness of the delta neutrophil index to predict 30-day mortality in patients with ST segment elevation myocardial infarction. Sci Rep. 2017;7: 15718. doi: 10.1038/s41598-017-15878-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Kong T, Park YS, Lee HS, Kim S, Lee JW, Yu G, et al. Value of the Delta Neutrophil Index for Predicting 28-Day Mortality in Patients With Acute Pulmonary Embolism in the Emergency Department. Shock. 2018;49: 649–657. doi: 10.1097/SHK.0000000000001027 . [DOI] [PubMed] [Google Scholar]
16.Yune HY, Chung SP, Park YS, Chung HS, Lee HS, Lee JW, et al. Delta neutrophil index as a promising prognostic marker in out of hospital cardiac arrest. PLoS One. 2015;10: e0120677. doi: 10.1371/journal.pone.0120677 . [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Kong T, Park YS, Lee HS, Kim S, Han S, Eun CH, et al. A Delta Neutrophil Index for the Prediction of Contrast-Induced Nephropathy in Patients With St-Elevation Myocardial Infarction Followed By Percutaneous Coronary Intervention. Shock. 2018;49: 317–325. doi: 10.1097/SHK.0000000000000957 . [DOI] [PubMed] [Google Scholar]
18.Kim H, Kong T, Chung SP, Hong JH, Lee JW, Joo Y, et al. Usefulness of the Delta Neutrophil Index as a Promising Prognostic Marker of Acute Cholangitis in Emergency Departments. Shock. 2017;47: 303–312. doi: 10.1097/SHK.0000000000000722 . [DOI] [PubMed] [Google Scholar]
19.Kim JH, Park YS, Yoon CY, Lee HS, Kim S, Lee JW, et al. Delta Neutrophil Index for the Prediction of the Development of Sepsis-Induced Acute Kidney Injury in the Emergency Department. Shock. 2019;52: 414–422. doi: 10.1097/SHK.0000000000001299 . [DOI] [PubMed] [Google Scholar]
20.Kong T, Park YS, Lee HS, Kim S, Lee JW, You JS, et al. The Delta Neutrophil Index Predicts the Development of In-hospital Hypotension in Initially Stable Patients with Pyogenic Liver Abscess. Sci Rep. 2019;9: 12105. doi: 10.1038/s41598-019-48588-1 . [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Park JH, Byeon HJ, Lee KH, Lee JW, Kronbichler A, Eisenhut M, et al. Delta neutrophil index (DNI) as a novel diagnostic and prognostic marker of infection: a systematic review and meta-analysis. Inflamm Res. 2017;66: 863–870. doi: 10.1007/s00011-017-1066-y . [DOI] [PubMed] [Google Scholar]
22.Manz MG, Boettcher S. Emergency granulopoiesis. Nat Rev Immunol. 2014;14: 302–314. doi: 10.1038/nri3660 . [DOI] [PubMed] [Google Scholar]
23.Cain DW, Snowden PB, Sempowski GD, Kelsoe G. Inflammation triggers emergency granulopoiesis through a density-dependent feedback mechanism. PLoS One. 2011;6: e19957. doi: 10.1371/journal.pone.0019957 . [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Leliefeld PH, Wessels CM, Leenen LP, Koenderman L, Pillay J. The role of neutrophils in immune dysfunction during severe inflammation. Crit Care. 2016;20: 73. doi: 10.1186/s13054-016-1250-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Alves-Filho JC, Spiller F, Cunha FQ. Neutrophil paralysis in sepsis. Shock. 2010;34 Suppl 1: 15–21. doi: 10.1097/SHK.0b013e3181e7e61b . [DOI] [PubMed] [Google Scholar]
26.Livingston DH, Anjaria D, Wu J, Hauser CJ, Chang V, Deitch EA, et al. Bone marrow failure following severe injury in humans. Ann Surg. 2003;238: 748–753. doi: 10.1097/01.sla.0000094441.38807.09 . [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Livingston DH, Gentile PS, Malangoni MA. Bone marrow failure after hemorrhagic shock. Circ Shock. 1990;30: 255–263 . [PubMed] [Google Scholar]
28.Guérin E, Orabona M, Raquil MA, Giraudeau B, Bellier R, Gibot S, et al. Circulating immature granulocytes with T-cell killing functions predict sepsis deterioration*. Crit Care Med. 2014;42: 2007–2018. doi: 10.1097/CCM.0000000000000344 . [DOI] [PubMed] [Google Scholar]
29.Figueiredo S, Taconet C, Harrois A, Hamada S, Gauss T, Raux M, et al. How useful are hemoglobin concentration and its variations to predict significant hemorrhage in the early phase of trauma? A multicentric cohort study. Ann Intensive Care. 2018;8: 76. doi: 10.1186/s13613-018-0420-8 . [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0258619.r001

Decision Letter 0

Wisit Cheungpasitporn

26 Apr 2021

PONE-D-20-31096

Delta neutrophil index and shock index can predict the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

PLOS ONE

Dear Dr. Je Sung You,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR: The reviewers have raised a number of points which we believe major modifications are necessary to improve the manuscript, taking into account the reviewers' remarks. Please consider and address each of the comments raised by the reviewers before resubmitting the manuscript. This letter should not be construed as implying acceptance, as a revised version will be subject to re-review.

==============================

Please submit your revised manuscript by Jun 10 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2) Thank you for stating the following in the Competing Interests section:

[NO The authors have declared that no competing interests exist.].

We note that you received funding from a commercial source: Siemens Health Care

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

Reviewer #4: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: 1. Abstract (background): The study purpose is not described.

2. Abstract (methods): “…to evaluate the usefulness of the DNI as an indicator of sterile inflammation severity” This is a vague description. Please indicate the purpose clearly.

3. Abstract (result: “Of the 278 patients enrolled, 60 required MT.” This sentence should be moved to methods section.

4. Keywords: “immature neutrophils” can be deleted.

5. Introduction: There are several definitions of PPH, including severe PPH, early PPH, and primary PPH. How can these definitions be applied to the study population? This information can be presented in Table 1 or Figure 1.

6. Introduction (page 6): Most of the paragraph is composed of lengthy, descriptive narration on DNI application in various clinical situations. This part can be shortened compactly.

7. Methods (page 7): How many patients were transferred from another hospital or a private clinic to ED? How many of them were transfused already before ED arrival? These baseline data should be presented in Table 1 or Figure 1.

8. Methods (page 9): During the 10-yr study period, was the same instrument (ADVIA 2120) used continuously without change?

9. Methods (page 10): In line with my comment #7, clinical endpoints “the amount of transfused blood before ED arrival and the amount of transfused blood after ED arrival” can be presented in Table 1 or Figure 1.

10. Statistical analysis: Although it is described that “The data are presented as medians and interquartile ranges for continuous variables”, the data are presented as mean/SD in Table 1.

11. Statistical analysis (page 11): diagnostic performance? diagnostic accuracy? Is it for the diagnosis of massive transfusion or for the prediction of massive transfusion? If it is for the prediction, throughout the text, predictive performance should be described instead of diagnostic performance.

12. The whole “RESULTS” section has lots of duplicated descriptions of each Table and Figure. Please avoid duplicated presentation. Moreover, the first subsection, “Patients’ characteristics”, should be moved to the Methods section. In this retrospective study, this part is not the result corresponding to the study purpose but the basic description on the study population.

13. Table 2 can be modified to show the univariate analysis results together. That would be much more comprehensive.

14. Fig. 2: Did the authors compare AUC values among DNI, SI, DNI + SI? Please provide the statistical data (pairwise comparison of AUC values). The comparison of 0.821 vs. 0.831 (with overlapping CIs) does not seem to provide a significant result.

15. Discussion section is quite lengthy spanning 4.5 pages (including Conclusion). It should be shortened compactly. The third paragraph (pages 14 – 16) is full of general descriptions that are not directly related to the study results. In the fourth paragraph (page 16), what was the lactate levels in this study population? If lactate levels are so important (as described in this paragraph), it deserves to be included in Table 1 and can be analyzed together.

16. This study used only the initial DNI values at ED admission. Follow-up DNI values would be available in these patients. Considering the time definition of massive transfusion, what was the DNI value and shock index value after 24 hrs?

17. In Table 1, how many patients had shock index > 1.0 in the two groups? The laboratory data should be presented in SI units.

18. In addition to the ROC curve analysis, IDI/NRI analyses are recommended to show additive value (any superiority) of DNI value to shock index.

Overall, the readability and organization of this manuscript is acceptable. However, the additive value of DNI value on top of shock index should be more emphasized to support their conclusion. Although the authors mentioned “screening tool” and “additional prediction tool” as the clinical usefulness of DNI value on top of shock index (page 17), which change in detail can we expect in the clinical practice? Can we decrease the transfused blood amount? Can we decrease the number of patients who were transfused massively? The authors performed a retrospective analysis; so, if they want to emphasize the adding value of DNI onto shock index, they can propose an algorithm (including both DNI value and shock index) for the clinical decision-making and can show the simulation results how the massive transfusion practice would have been changed.

Reviewer #2: The authors demonstrate a complex statistical association to integrate into daily medical practice. Moreover, the primary and secondary objectives are not at all clear. The method lacks precision; it is a study with a change of initial destination, but what was the purpose of the first study? what were the criteria?

The sample is small and few values are really significant.

This article does not provide any real benefit to the medical literature

Reviewer #3: Dear Editor,

Thanks for the opportunity to read the article.

It was a well-written and well-designed study that should be accepted for publication.

My recommendation is to accept the article in its current form.

Best regards.

Reviewer #4: The manuscript lines are not numbered, so I identified the lines of the manuscript by quoting parts of them.

Introduction

“…is the cause of maternal mortality in 12% of cases…” I assume what you mean is that 12% of maternal mortality is caused by PPH. Is that correct?

“An SI of >1.0 showed a specificity of 78.7%, a sensitivity of 59.2%, a positive predictive value

(PPV) of 58.7%, and a negative predictive value (NPV) of 79.1%.” Predictive values, although often reported, are useless statistics because they change with the prevalence of disease and you are trying to estimate the prevalence of disease in the individual patient. predictive values have almost no clinical usefulness, so please remove them from the paper. Although sensitivities and specificities are somewhat useful, likelihood ratios are more useful since they can be applied to the individual patient. Therefore, better would be a statement such as, “Using the results from reference 7, for a SI > 1.0, the positive and negative likelihood ratios for massive transfusion in patients with PPH are 3 and the negative likelihood ratio is 0.5, making SI of little help to the clinician for predicting the need for transfusion.” In general likelihood ratios are useful only if they are >5 or <0.2.

Methods

“…and identified…PPV, and NPV…” As mentioned above, PPV and NPV are useless statistics. Please eliminate them from the paper.

Results

“…SBP (94.267…). Reporting SBP to 3 significant figures is not needed. Better to report the results for blood pressures and heart rates as whole numbers. This will also make the paper easier to read.

I request two general changes in the paper. One has to do with the presentation of the results. As mentioned above, please remove predictive values and include likelihood ratios in your results. The second problem is the interpretation of the ROC curves as described next.

Fig. 2 The ROC curves for the DNI and DNI+SI are almost overlapping so it is slightly surprising that your sensitivities and specificities (Table 3) are so different. I think this is most likely because you have chosen different points on the curves to calculate them. I have described the method I use below, that I think is preferable. Nonetheless, the likelihood ratios calculated from the sensitivities and specificities listed in Table 3 are shown in the following table. Using these likelihood ratios, I have also calculated the post-test probabilities for need for massive blood transfusion, assuming that the pretest probability is 21.6% (60/278).

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

DNI 3.5 0.48 49% 11%

SI 3.3 0.31 47% 8%

DNI + SI 11.6 0.49 76% 12%

The pretest probability of MT requirement is 21.6%. The question is, does applying DNI and/or SI change that significantly enough to benefit the clinician seeing the next patient that may require MT (and, if so, what does the clinician do differently than would not have been done if this information was not available)? As you can see, adding DNI to SI is not helpful if the test is negative, but somewhat helpful if the test is positive. In general, applying the results of your study to patients with PPH would reduce the probability of requiring MT in patients with a negative test by half, yet 10% of those patients still require MT. On the other hand, of those with a positive test, about one quarter will end up not requiring MT, using the DNI+SI parameters. Please comment on how this would help the clinician actually caring for these patients, in particular, once this result is obtained, how does the plan of the physician change? Would the clinician then immediately initiate the massive transfusion protocol, knowing that about one-quarter of the patients do not need it? I think that this type of analysis points out that even when differences are highly statistically significant, this does not necessary translate into a result that is that useful to the clinician seeing the next patient with PPH.

I performed two other calculations on your data. First, I had some difficulty with data taken from the DNI+SI ROC graph, since the point on the graph where sensitivity = 53.5% and specificity = 95.4% is not actually on the line. However, the point on the SI and SI+DNI lines closest to this point show the following results:

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

SI 5.6 0.54 60% 13%

DNI + SI 7.3 0.53 67% 13%

This suggests that adding DNI to the SI is only marginally helpful. This is particularly important for two reasons: 1. The SI is immediately available once vital signs are taken and 2. Many clinicians may not be familiar with DNI and/or may have difficulty remembering the formula to calculate it or be able to calculate it themselves.

I actually calculated the same values by finding the points on the curves in Fig. 2 that are closest to the left upper corner of the graph (where sensitivity and specificity = 1). I did this by printing the graph and using a compass to find the point. This produced the following table, corresponding to the one above.

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

DNI 2.8 0.45 43% 11%

SI 3.8 0.30 41% 7%

DNI+SI 5.0 0.35 57% 9%

As you can see, the results are similar to yours except for the DNI+SI with a positive result. This also suggests that adding DNI to SI is only a little helpful If the test is positive, but not at all if the test is negative.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: Yes: Barnet Eskin

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Oct 15;16(10):e0258619. doi: 10.1371/journal.pone.0258619.r002

Author response to Decision Letter 0

17 Jul 2021

Reviewer #1:

1. Abstract (background): The study purpose is not described.

Response: Thank you for your comment. We agree with you. We have revised the abstract accordingly, as follows:

“Background: Postpartum hemorrhage (PPH) constitutes a major risk for maternal mortality and morbidity. Unfortunately, the severity of PPH can be underestimated because it is difficult to accurately measure blood loss by visual estimation. The delta neutrophil index (DNI), which reflects the concentration of circulating immature granulocytes, is automatically calculated in hematological analyzers. We evaluated the significance of the DNI in predicting hemorrhage severity based on the requirement for massive transfusion (MT) in patients with PPH.”

2. Abstract (methods): “…to evaluate the usefulness of the DNI as an indicator of sterile inflammation severity” This is a vague description. Please indicate the purpose clearly.

Response: Thank you for your comment. We agree with your comment. We have revised the abstract accordingly, as follows:

“Methods: We retrospectively analyzed data from a prospective registry to evaluate the association between the DNI and MT. Moreover, we assessed the predictive ability of the combination of DNI and shock index (SI) for the requirement for MT. MT was defined as a transfusion of ≥10 units of red blood cells within 24 h of PPH. In total, 278 patients were enrolled in this study and 60 required MT.”

3. Abstract (result: “Of the 278 patients enrolled, 60 required MT.” This sentence should be moved to methods section.

Response: Thank you for your comment. We agree with your suggestion. We have revised the abstract accordingly, as shown above in the response to Comment No. 2.

4. Keywords: “immature neutrophils” can be deleted.

Response: Thank you for your comment. In line with your recommendation, we have removed the term “immature neutrophils” from the list of keywords.

Response: Thank you for your comment.

You have pointed out that patients enrolled in this study should be classified according to the three definitions of PPH: severe, early, and primary. In the Introduction section, we have presented the general characteristics and various definitions of PPH in relation to various related studies.

PPH is defined as a blood loss of ≥ 500 mL, and severe PPH is defined as a blood loss of ≥ 1,000 mL. According to the American College of Obstetricians and Gynecologists, early PPH is defined as a total blood loss of at least 1,000 mL or blood loss with symptoms and signs of hypovolemia within 24 h after intrapartum loss or delivery of fetus [1,4]. Primary PPH may develop before placenta delivery and up to 24 h after fetus delivery [1]. In the present study, all enrolled patients had primary PPH and experienced bleeding within 24 h after fetus delivery. In addition, since all patients were transferred from another hospital or a private clinic due to severe bleeding, it can be considered that they all met the criteria for severe and early PPH.

To accurately classify PPH as severe or early based on the above definitions, it is necessary to estimate the amount of blood loss. However, as suggested in previous studies, it is difficult to accurately estimate the amount of blood loss in clinical settings. In our study, since all patients were transferred from another hospital or a private clinic due to severe bleeding, this would be very difficult. Moreover, we believe that classifying patients based on retrospectively estimated amount of blood loss may lead to erroneous interpretation of the study results.

The purpose of this study was to indirectly estimate the amount of blood loss in PPH, which is difficult to accurately measure, using the DNI and DNI+SI and predict the requirement for MT. Therefore, it is difficult to accurately classify patients based on these definitions of PPH.

6. Introduction (page 6): Most of the paragraph is composed of lengthy, descriptive narration on DNI application in various clinical situations. This part can be shortened compactly.

Response: Thank you for your comment. As we agreed with your comment, we have revised the Introduction section as follows:

An SI of >1.0 showed a positive likelihood ratio of 2.8 and a negative likelihood ratio of 0.5 for the requirement for massive transfusion in patients with PPH, indicating that the SI is of little help to clinicians for predicting the need for transfusion [7]. In general, likelihood ratios are only useful if their values are >5 or <0.2.

However, additional biomarkers or indicators are needed to better predict the requirement for massive transfusion in patients with PPH in the emergency department (ED). With the advancement in technology, the delta neutrophil index (DNI) can be obtained using automated hematological analyzers based on differential leukocyte counts obtained from two independent channels[10,11]. The DNI, which reflects the proportion of circulating immature granulocytes, is automatically calculated as the difference between the leukocyte differentials measured in the myeloperoxidase channel and those detected in the lobularity channel[10,11]. Many recent studies have demonstrated that a higher DNI indicates higher severity in infectious diseases such as suspected sepsis, septic shock, urinary tract infection, acute appendicitis, and liver abscess[11-15]. Many recent studies have founded that the DNI was significantly associated with the severity of diseases related to sterile inflammation, such as out-of-hospital cardiac arrest, major traumatic injury, pulmonary embolism, acute myocardial infarction, and upper gastrointestinal hemorrhage[10,16-19]. As an inflammatory response to hypoperfusion and reperfusion owing to blood loss and tissue damage can develop into systemic inflammatory response syndrome, the DNI is a promising candidate for a biomarker for predicting the requirement of massive transfusion in patients with primary PPH in the ED[10,16]. In particular, the DNI can be useful in emergency settings because it can be rapidly and easily determined using an automated blood cell analyzer while measuring the complete blood count (CBC)[10,11,16].

Response: Thank you for your comment. This study enrolled only patients who were referred to our ED for evaluation and management of PPH from other hospitals or private obstetric clinics within the first 24 h of delivery. To determine the amount of transfusion, we measured both the amount of transfused blood before arrival at the ED and that after arrival at the ED (in the ED and intensive care unit [ICU] or general ward). We have added data for the patients who received blood transfusion before arrival at the ED in Table 1.

8. Methods (page 9): During the 10-yr study period, was the same instrument (ADVIA 2120) used continuously without change?

Response: Thank you for your comment. We have checked the information about the equipment. Our institution has consistently used the same equipment in the ED during the study period for research purposes. Replacement was made with an upgraded version of the same equipment (ADVIA 2120 -> ADVIA 2120i).

Response: Thank you for your comment. Accordingly, we have added data on the amount of transfused blood before and after arrival at the ED in Table 1.

10. Statistical analysis: Although it is described that “The data are presented as medians and interquartile ranges for continuous variables”, the data are presented as mean/SD in Table 1.

Response: Thank you for your comment. We have revised the manuscript accordingly, as follows:

“The data are presented as mean (SD) for continuous variables and as absolute or relative frequencies for categorical variables.”

Response: Thank you for your comment. We agree with your comment. Accordingly, we have revised “diagnostic” to “predictive” in the manuscript.

Response: Thank you for your comment. We agree with you. Accordingly, we have moved the relevant text to the Methods section of the manuscript.

13. Table 2 can be modified to show the univariate analysis results together. That would be much more comprehensive.

Response: Thank you for your comment. We agree with your suggestion. We have revised Table 2 to show the univariate analysis results together.

Response: Thank you for your comment. We agree with your comment. Accordingly, we have added pairwise comparison of the AUC values to supplement our results. We discussed this with our statisticians. We found that there was no significant difference between the SI and DNI+SI using pairwise comparison of the AUC values. However, the DNI+SI was significantly superior to the SI in the Net Reclassification Index (NRI) and Integrated Discrimination Index (IDI) analyses (see the response to reviewer’s comment no. 18). It is thought that this difference in results can be attributed to the AUC comparison, which has a more conservative tendency compared with the NRI or IDI.

Response: Thank you for your comment. We agree with your comment. We have reduced the Discussion section accordingly, as shown below. In addition, we have added the lactate levels in Table 1.

“The DNI is available in emergency settings because it can be rapidly and easily determined using an automated blood cell analyzer while measuring the CBC[10,11,16] The DNI value is correlated with the manual count of immature granulocytes[11]. However, automatic calculation of the DNI is a more prompt and accurate method than manual counting of immature granulocytes[10]. The diagnostic criteria for systemic inflammatory response syndrome include an increase in the number of immature granulocytes in the circulation as an important factor[10,16]. Several studies have shown that an increase in the DNI can predict the severity of infection and sterile inflammation without infection and that it is significantly associated with poor outcomes in specific disease conditions[10,13-18]. Massive hemorrhage causes detrimental systemic inflammation without infection[10,16]. The sustained, exacerbated inflammatory response is strongly associated with increased mortality[10,16]. Emergency granulopoiesis is the most accepted theory for explaining why an elevated DNI value is associated with the use of massive transfusion after primary PPH[21,22]. Emergency granulopoiesis implies the expansion of the immature granulocytic population for providing hematopoietic precursors to sustain the increasing number of circulating neutrophils[10,21,22]. The number of circulating immature neutrophils can be rapidly increased to compensate for the destruction of mature neutrophils due to severe systemic inflammation and the loss of active neutrophils due to massive consumption related to sterile stimuli, such as chemical agents, trauma, bleeding, and ischemic insult[21,23,24]. Livingston et al. proposed that bone marrow failure can develop due to insufficient perfusion and ischemia during hemorrhagic shock[25,26]. This bone marrow dysfunction causes a transient failure of granulopoiesis[25,26]. Consequently, severe hemorrhage may contribute to the release of immature granulocytes into the circulation[23,27]. As pregnant women often have prenatal anemia, it is difficult to determine the occurrence of massive hemorrhage and evaluate the requirement for massive transfusion based on hemoglobin levels alone in patients with PPH[28]. Although lower hemoglobin or hematocrit levels have been widely and interchangeably used as indicators of severe bleeding, their diagnostic values for determining the severity of blood loss in the initial phase in patients with severe hemorrhage, such as trauma-related and GI bleeding, remain controversial[28]. In the present study, the hemoglobin level on ED admission was not associated with the requirement for massive transfusion.

In addition, some studies have reported that the initial SI value, as a hemodynamic indicator based on the HR and SBP (easily obtained in the ED), could predict the requirement for massive transfusion. Although the initial SI value on ED admission was independently associated with the requirement for massive transfusion in patients with primary PPH, it had a relatively low specificity[7,8]. Sohn et al. revealed that combining the initial SI value with the lactate level could improve its predictive performance for the requirement for massive transfusion in patients with primary PPH[7]. As increased plasma lactate levels are significantly associated with increased morbidity and mortality, lactate measurement is a valuable tool for triage and risk stratification. Hyperlactatemia is often attributed to tissue hypoxia in several diseases[29]. Conversely, an increased plasma lactate level might not always signify hypoxia, thus reducing its specificity[30]. Hyperlactatemia without hypoxia might result from aerobic glycolysis during sepsis, liver failure, altered pyruvate dehydrogenase activity, hyperventilation, and elevated catecholamine levels; it might also occur in the presence of certain drugs and toxins[31-33]. This may negatively affect the usefulness of lactate for predicting disease severity.”

Response: Thank you for your helpful comment. In line with your comment, we have analyzed the DNI and SI values at 24 h after ED admission. In these results, the DNI value was significantly higher in patients with requirement for MT than in those with no requirement for MT. However, there was no significant difference in the SI at 24 h after ED admission between the two groups.

17. In Table 1, how many patients had shock index > 1.0 in the two groups? The laboratory data should be presented in SI units.

Response: Thank you for your comment. We have revised Table 1 accordingly. Among all patients, 94 (34%) had an SI >1.0. There were significantly more patients with an SI >1.0 among those with requirement for MT than among those with no requirement for MT.

18. In addition to the ROC curve analysis, IDI/NRI analyses are recommended to show additive value (any superiority) of DNI value to shock index.

Response: Thank you for your comment. We agree with your comment. Accordingly, we have added pairwise comparison of the AUC values to supplement our findings. We found that there was no significant difference between the SI and DNI+SI using pairwise comparison of the AUC values. However, the DNI+SI was significantly superior to the SI in the NRI and IDI analyses. It is thought that this difference in results is caused by the AUC comparison, which has a more conservative tendency compared with the NRI or IDI.

Overall, the readability and organization of this manuscript is acceptable.

However, the additive value of DNI value on top of shock index should be more emphasized to support their conclusion. Although the authors mentioned “screening tool” and “additional prediction tool” as the clinical usefulness of DNI value on top of shock index (page 17), which change in detail can we expect in the clinical practice? Can we decrease the transfused blood amount? Can we decrease the number of patients who were transfused massively? The authors performed a retrospective analysis; so, if they want to emphasize the adding value of DNI onto shock index, they can propose an algorithm (including both DNI value and shock index) for the clinical decision-making and can show the simulation results how the massive transfusion practice would have been changed.

Response: Thank you for your helpful comment. We have discussed this with our statisticians. In this study, we demonstrated that the DNI was independently associated with the requirement for MT in patients with PPH. In comparison with the SI alone, the addition of the DNI to the SI showed significant increases in specificity, accuracy, LR+, and post-test probability if the test was positive. However, there was no improvement in sensitivity, LR-, and post-test probability if the test was negative. Nevertheless, we believe that the application of the DNI has its own clinical value in improving the prediction of hemorrhage severity and requirement for MT in patients with PPH.

In this study, MT was defined as a transfusion of ≥ 10 units of packed red blood cells. MT is not a simple medical practice that is decided based on the initial condition of patients in the ED; it requires monitoring of laboratory results and vital signs and medical judgment based on the changes of patient’s condition during the critical period.

As resources like blood tests and ICU are limited in the ED, it is important to indirectly estimate the severity of the hemorrhage, the probability of MT requirement, or the actual amount of bleeding. Considering the limited medical resources, it is of great importance to improve the specificity of prediction for the requirement for MT in the ED.

This study enrolled only patients who were referred to our ED for evaluation and management of PPH from other hospitals or private obstetric clinics within the first 24 h of delivery. Basically, we applied standardized protocols (SPEED CP) to these patients and evaluated patients with PPH regarding requirement for high-level monitoring and MT. For these patients, we believe that it is clinically meaningful to screen for the need for MT accurately using the DNI value obtained through a simple CBC test. It is difficult to clearly determine the requirement for MT in patients with PPH using only the DNI and DNI+SI values; however, we believe that these would be of great clinical utility as auxiliary indicators in patients with PPH. Of course, it is unfortunate that the sensitivity, LR-, and post-test probability in cases of negative tests did not improve after adding the DNI to the SI. Further study is needed to clarify the advantages and disadvantages.

Regarding the algorithm suggested by the reviewer, the DNI+SI can clearly increase the likelihood ratio for requirement for MT. However, clinical decision making for MT requirement is a very complex process that combines assessments drawn by monitoring of laboratory results and vital signs and medical judgment based on the changes of patient’s condition over a considerable period of time. Therefore, we think that establishing an algorithm related to MT requirement determination using only this value may confuse readers. Rather than determining the requirement for MT in patients with PPH based on the DNI and DNI+SI values alone, it would be appropriate to use them as auxiliary indicators. Based on your comment, we have added this point in the manuscript, as follows:

“Although clinical decision making for MT requirement is a very complex process in the ED, further study is needed to establish a clinical algorithm for determining the requirement for MT in patients with PPH based on the DNI and DNI+SI values alone.”

Reviewer #2:

The authors demonstrate a complex statistical association to integrate into daily medical practice. Moreover, the primary and secondary objectives are not at all clear. The method lacks precision; it is a study with a change of initial destination, but what was the purpose of the first study? what were the criteria? The sample is small and few values are really significant. This article does not provide any real benefit to the medical literature

Response: Thank you for your helpful comment. We understand that you have raised questions about the purpose and clinical applicability of this study. Considering the fact that there are not enough useful predictors of hemorrhage severity in patients with PPH, this study investigated the possibility of predicting the requirement for MT using the DNI, which has recently been found useful in predicting the severity of sterile inflammatory disease (out of hospital cardiac arrest, trauma, ST elevation MI, and upper GI bleeding).

In this study, we demonstrated that the DNI was independently associated with the requirement for massive transfusion in patients with PPH. In comparison with the SI alone, the addition of the DNI to the SI showed significant increases in the specificity, accuracy, LR+, and post-test probability if the test was positive. However, there was no improvement in sensitivity, LR-, and post-test probability if the test was negative. Nevertheless, we believe that the application of the DNI has its own clinical value in improving the prediction of hemorrhage severity and requirement for MT in patients with PPH.

In this study, MT was defined as a transfusion of ≥ 10 units of packed red blood cells. MT is not a simple medical practice that is decided based on the initial condition of patients in the ED, but it is a medical practice that combines monitoring of laboratory results and vital signs and medical judgment based on the changes of patient’s condition during the critical period.

As resources like blood tests and ICU are limited in the ED, it is important to indirectly estimate the severity of hemorrhage, the probability of MT requirement, or the actual amount of bleeding. Considering the limited medical resources, it is of great importance to improve the specificity of MT prediction in the ED.

Reviewer #3:

Dear Editor,

Thanks for the opportunity to read the article.

It was a well-written and well-designed study that should be accepted for publication.

My recommendation is to accept the article in its current form.

Best regards.

Response: We greatly appreciate your comments.

Reviewer #4:

The manuscript lines are not numbered, so I identified the lines of the manuscript by quoting parts of them.

Response: We apologize for the inconvenience. We have revised the resubmitted manuscript accordingly.

Introduction

“…is the cause of maternal mortality in 12% of cases…” I assume what you mean is that 12% of maternal mortality is caused by PPH. Is that correct?

Response: Thank you for pointing this out. We apologize for the confusion. We have revised the text accordingly, as follows:

“PPH constitutes a major risk for maternal mortality and morbidity in both developed and developing countries and is the cause of maternal mortality in 12% of cases in the United States.

PPH constitutes a major risk for maternal mortality and morbidity in both developed and developing countries and causes 12% of maternal mortality in the United States.”

“An SI of >1.0 showed a specificity of 78.7%, a sensitivity of 59.2%, a positive predictive value (PPV) of 58.7%, and a negative predictive value (NPV) of 79.1%.” Predictive values, although often reported, are useless statistics because they change with the prevalence of disease and you are trying to estimate the prevalence of disease in the individual patient. predictive values have almost no clinical usefulness, so please remove them from the paper.

Response: Thank you for your helpful comment. We agree with your comment. We have deleted the statistics on PPV and NPV in the revised manuscript and Tables and added likelihood ratios in our results.

Although sensitivities and specificities are somewhat useful, likelihood ratios are more useful since they can be applied to the individual patient. Therefore, better would be a statement such as, “Using the results from reference 7, for a SI > 1.0, the positive and negative likelihood ratios for massive transfusion in patients with PPH are 3 and the negative likelihood ratio is 0.5, making SI of little help to the clinician for predicting the need for transfusion.” In general likelihood ratios are useful only if they are >5 or <0.2.

Response: Thank you for your helpful comment. We agree with your comment. We have revised the manuscript accordingly, as follows:

“An SI of >1.0 showed a positive likelihood ratio of 2.8 and a negative likelihood ratio of 0.5 for the requirement for massive transfusion in patients with PPH, indicating that the SI is of little help to clinicians for predicting the need for transfusion [7]. In general, likelihood ratios are only useful if their values are >5 or <0.2.”

Methods

“…and identified…PPV, and NPV…” As mentioned above, PPV and NPV are useless statistics. Please eliminate them from the paper.

Response: Thank you for your helpful comment. As per your suggestion, we have deleted the statistics on PPV and NPV from the revised manuscript.

Results

Response: Thank you for your helpful comment. Accordingly, we have revised the systolic blood pressure and heart rate values as whole numbers in the revised manuscript.

I request two general changes in the paper. One has to do with the presentation of the results. As mentioned above, please remove predictive values and include likelihood ratios in your results.

Response: Thank you for your helpful comment. We have deleted the statistics on PPV and NPV in the revised manuscript and Tables and added likelihood ratios in our results.

The second problem is the interpretation of the ROC curves as described next.

I have described the method I use below, that I think is preferable. Nonetheless, the likelihood ratios calculated from the sensitivities and specificities listed in Table 3 are shown in the following table. Using these likelihood ratios, I have also calculated the post-test probabilities for need for massive blood transfusion, assuming that the pretest probability is 21.6% (60/278).

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

DNI 3.5 0.48 49% 11%

SI 3.3 0.31 47% 8%

DNI + SI 11.6 0.49 76% 12%

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

SI 5.6 0.54 60% 13%

DNI + SI 7.3 0.53 67% 13%

LR+ LR- Post-test probability if test positive Post-test probability if test is negative

DNI 2.8 0.45 43% 11%

SI 3.8 0.30 41% 7%

DNI+SI 5.0 0.35 57% 9%

Response: We are extremely grateful for the time you have taken to review our data. In this study, we demonstrated that the DNI was independently associated with the requirement for massive transfusion in patients with PPH. In comparison with the SI alone, the addition of the DNI to the SI showed significant increases in specificity, accuracy, LR+, and post-test probability if the test was positive. However, there was no improvement in sensitivity, LR-, and post-test probability if the test was negative. Nevertheless, we believe that the application of the DNI has its own clinical value to improve the prediction of hemorrhage severity and requirement for massive transfusion in patients with PPH.

In this study, MT was defined as a transfusion of ≥ 10 units of packed red blood cells. MT is not a simple medical practice that is decided based on the initial condition of patients in the ED, but it is a medical practice that combines monitoring of laboratory results and vital signs and medical judgment based on the changes of patient’s condition over a considerable period of time.

Attachment

Submitted filename: renamed_dbaff.docx

Click here for additional data file.^{(116.4KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0258619.r003

Decision Letter 1

Wisit Cheungpasitporn

9 Aug 2021

PONE-D-20-31096R1

Delta neutrophil index and shock index can predict the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

PLOS ONE

Dear Dr. You,

ACADEMIC EDITOR: Our expert reviewers still have raised a number of points which we believe major modifications are necessary to improve the revised manuscript, taking into account the reviewers' remarks below.

Please submit your revised manuscript by Sep 23 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #4: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #4: No

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #4: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #4: Yes

**********

6. Review Comments to the Author

Although the information in the “Abstract” is correct, it is of little interest to the clinician. That the odds ratios are statistically significant is of little help to the clinician about to care for the next patient with PPH. It would be better to substitute information about likelihood ratios since they can be used directly by the clinician to take care of the patient. Even if the authors choose to retain odds ratios in the abstract, they should also include the information that values below the cutoff were not helpful, whether they were statistically significant or not.

I had previously made the comment about the number of significant figures: “…SBP (94.267…). Reporting SBP to 3 significant figures is not needed. Better to report the results for blood pressures and heart rates as whole numbers. This will also make the paper easier to read.

The response was: “Thank you for your helpful comment. Accordingly, we have revised the systolic blood pressure and heart rate values as whole numbers in the revised manuscript.” However, I did not see where this was actually done. Please follow up with this. Furthermore, Table 1 would be easier to read if values were rounded out to whole numbers. This would not lose anything in terms of conclusions reached by the study. For example, the first entry, 32.70±3.95 is better reported as 33 ± 4. There are other places in the body of the paper where results are given with an inordinately large number of significant figures. This just makes the paper harder to read, does not add anything to the interpretation of the results and is actually not justified from a statistical point-of-view. For example, “…with a specificity of 83.03% (95% CI: 78.05–88.01)” implies that the specificity of 83.04 is incorrect, whereas the 95% CI includes that value. I would be better to say, “…with a specificity of 83% (95% CI: 78-88).” Furthermore, I really don’t think anyone cares whether the specificity is 83.03% or 83%.

Regarding the following sentence: “On combining a DNI value of ≥3.3 with an SI value of ≥1.0, the specificity and PLR increased to 95.41% (95% CI; 92.63–98.19) and 11.63 (95% CI: 6.07–22.27), respectively.” I had some difficulty with data taken from the DNI+SI ROC graph, since the point on the graph where sensitivity = 53.5% and specificity = 95.4% is not actually on the line (see graph in attachment, point A). The authors did not address this concern in their response to the initial review even though this is crucial for interpreting the data. Their result (PLR for DNI+SI = 11.63) is much higher than the 7.3 or 5.0 that I calculated using points that are actually on the plotted line (attached graph, points B and C). The authors need to address this since you cannot arbitrarily choose any point on the graph to make the calculations. You need to use a point that is actually on the line in the graph. Please make this correction in the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #4: No

Attachment

Submitted filename: ROC - 8.3.2021.pdf

Click here for additional data file.^{(264.5KB, pdf)}

PLoS One. 2021 Oct 15;16(10):e0258619. doi: 10.1371/journal.pone.0258619.r004

Author response to Decision Letter 1

29 Sep 2021

Please, check additional file for review resposne to reviewer's comments.

*** Response to reviewer’ s comments ***

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Response to reviewer #4’s comments

Reviewer #4: Title: “Delta neutrophil index and shock index can predict the requirement…” I think the use of the word “predict” offers something to the readers that the study does not actually provide. Perhaps better would be, “Delta neutrophil index and shock index can risk stratify the requirement…” Although the information in the “Abstract” is correct, it is of little interest to the clinician. That the odds ratios are statistically significant is of little help to the clinician about to care for the next patient with PPH. It would be better to substitute information about likelihood ratios since they can be used directly by the clinician to take care of the patient. Even if the authors choose to retain odds ratios in the abstract, they should also include the information that values below the cutoff were not helpful, whether they were statistically significant or not.

Response: Thank you for your astute comments and valuable suggestions. We agree with the reviewer’s opinion.

First, we have revised the title.

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Second, we have substituted odds ratios with likelihood ratios.

Results: Multivariable logistic regression revealed that the DNI and SI were independent predictors of MT. The optimal cut-off values of ≥3.3% and ≥1.0 for the DNI and SI, respectively, were significantly associated with an increased risk of MT (DNI: positive likelihood ratio [PLR] 3.54, 95% confidence interval [CI] 2.5–5.1 and negative likelihood ratio [NLR] 0.48, 95% CI 0.4–0.7; SI: PLR 3.21, 95% CI 2.4–4.2 and NLR 0.31, 95% CI 0.19–0.49). The optimal cut-off point for predicted probability was calculated for combining the DNI value and SI value with the equation derived from logistic regression analysis. Compared with DNI or SI alone, the combination of DNI and SI significantly improved the specificity, accuracy, and positive likelihood ratio of the MT risk.

Response: Thank you for your pertinent comments. We sincerely apologize for our oversight in the previous manuscript version. We have revised the values to whole numbers in this manuscript version.

Response: Thank you for your remarkable comments and insightful suggestions. As you mentioned, on combining a DNI value of ≥ 3.3 with an SI value of ≥ 1.0, an optimal cut-off point could not be represented using ROC curves. Therefore, we determined arbitrary point as an optimal cut-off point by combining the optimal cut-off points of both DNI value of ≥ 3.3 and SI value of ≥ 1.0. We believe that this optimal cut-off point can help in the clinical decision-making. However, we agree that we should measure sensitivity, specificity, and likelihood ratio based on the optimal cut-off point obtained on combining the DNI value and SI value represented on the ROC curve. We discussed this point with our statisticians and added the optimal cut-off point to the ROC curve using the equation derived from logistic regression.

To stratify risk for the requirement for massive transfusion by combining the DNI value and SI value in patients with primary postpartum hemorrhage in the ED, we used the following equation derived from logistic regression.

Predicted probability = 1/(1+exp (5.348-0.2464*DNI-3.2691.*SI)

Based on this equation, we calculated predicted probability for combining the DNI value and SI value. We determined the optimal cut-off point for predicted probability (DNI+SI) to be ≥ 0.249.

When we stratified risk for the requirement for massive transfusion with the predicted probability (≥ 0.249) by combining the DNI value and SI value in patients with primary postpartum hemorrhage in the ED, the application of the value significantly improved the specificity, sensitivity, and PLR for massive transfusion in comparison with SI alone. In addition, it also improved PLR and NLR in comparison with DNI alone. We have marked a “cross” on the ROC curve of Figure 2. This point represents the optimal cut-off point based on the predicted probability (DNI+SI) of ≥ 0.249.

We have deleted texts pertaining to the optimal cut-off point derived by combining the optimal cut-off points of both DNI value of ≥3.3 and SI value of ≥1.0 from the manuscript according to reviewer’s comment. We have added the optimal cut-off point to the ROC curve using the equation derived from logistic regression to the revised manuscript.

Method

In addition, to stratify risk for the requirement for massive transfusion by combining the DNI value and SI value, we used the equation derived from logistic regression analysis. Based on this equation, we calculated the predicted probability for combining the DNI value and SI value. We determined optimal the cut-off point for the predicted probability. Based on this optimal cut-off point, we measured sensitivity, specificity, PLR, NLR, and accuracy.

Results

The predicted probability derived from the logistic equation was computed for each patient. When the predicted probability of the equation [Predicted probability = 1/(1+exp(5.348-0.2464*DNI-3.2691.*SI)] containing the combination of the DNI and SI was more than 0.249, the specificity was 87% (95% CI: 82–91), sensitivity was 70% (95% CI: 58–82), accuracy was 83% (95% CI: 79–88), PLR was 5.26 (95% CI: 3.6–7.7), and NLR was 0.35 (95% CI: 0.2–0.5) for MT. The combination of the DNI and SI significantly improved the specificity (DNI vs DNI + S, p<0.001; SI ≥1.1 vs DNI + SI, p<0.001), accuracy (DNI vs DNI + SI, p<0.001; SI ≥1.1 vs DNI + SI, p<0.001), and PLR (DNI vs DNI + S, p<0.001; SI ≥1.1 vs DNI + SI, p=0.038; using the GEE method) for the requirement for massive transfusion when compared with either DNI or SI (Table 3).

Attachment

Submitted filename: Response to reviewer.docx

Click here for additional data file.^{(187.7KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0258619.r005

Decision Letter 2

Wisit Cheungpasitporn

4 Oct 2021

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

PONE-D-20-31096R2

Dear Dr. You,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Wisit Cheungpasitporn, MD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

I reviewed the revised manuscript and the response to reviewers' comments. Revised Manuscript is well written. All comments have been addressed and thus accepted for publication.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #4: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #4: Yes

**********

6. Review Comments to the Author

Reviewer #4: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #4: No

PLoS One. doi: 10.1371/journal.pone.0258619.r006

Acceptance letter

Wisit Cheungpasitporn

8 Oct 2021

PONE-D-20-31096R2

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Dear Dr. You:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Wisit Cheungpasitporn

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Univariate logistic regression analysis for clinical factors associated with the requirement for massive transfusion in patients with primary postpartum hemorrhage.

(DOCX)

Click here for additional data file.^{(18.7KB, docx)}

S2 Table. Original data that was used in the study.

(XLSX)

Click here for additional data file.^{(149.1KB, xlsx)}

Attachment

Submitted filename: renamed_dbaff.docx

Click here for additional data file.^{(116.4KB, docx)}

Attachment

Submitted filename: ROC - 8.3.2021.pdf

Click here for additional data file.^{(264.5KB, pdf)}

Attachment

Submitted filename: Response to reviewer.docx

Click here for additional data file.^{(187.7KB, docx)}

Data Availability Statement

All relevant data are within the manuscript and its Supporting Information files.

[pone.0258619.ref001] 1.Evensen A, Anderson JM, Fontaine P. Postpartum Hemorrhage: Prevention and Treatment. Am Fam Physician. 2017;95: 442–449 . [PubMed] [Google Scholar]

[pone.0258619.ref002] 2.Cho HY, Na S, Kim MD, Park I, Kim HO, Kim YH, et al. Implementation of a multidisciplinary clinical pathway for the management of postpartum hemorrhage: a retrospective study. Int J Qual Health Care. 2015;27: 459–465. doi: 10.1093/intqhc/mzv068 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref003] 3.Magann EF, Evans S, Hutchinson M, Collins R, Lanneau G, Morrison JC. Postpartum hemorrhage after cesarean delivery: an analysis of risk factors. South Med J. 2005;98: 681–685. doi: 10.1097/01.SMJ.0000163309.53317.B8 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref004] 4.Girault A, Deneux-Tharaux C, Sentilhes L, Maillard F, Goffinet F. Undiagnosed abnormal postpartum blood loss: Incidence and risk factors. PLoS One. 2018;13: e0190845. doi: 10.1371/journal.pone.0190845 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref005] 5.Rath WH. Postpartum hemorrhage—update on problems of definitions and diagnosis. Acta Obstet Gynecol Scand. 2011;90: 421–428. doi: 10.1111/j.1600-0412.2011.01107.x . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref006] 6.Sohn CH, Kim SR, Kim YJ, Seo DW, Ahn S, Lee YS, et al. Disseminated Intravascular Coagulation in Emergency Department Patients With Primary Postpartum Hemorrhage. Shock. 2017;48: 329–332. doi: 10.1097/SHK.0000000000000852 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref007] 7.Sohn CH, Kim YJ, Seo DW, Won HS, Shim JY, Lim KS, et al. Blood lactate concentration and shock index associated with massive transfusion in emergency department patients with primary postpartum haemorrhage. Br J Anaesth. 2018;121: 378–383. doi: 10.1016/j.bja.2018.04.039 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref008] 8.Sohn CH, Kim WY, Kim SR, Seo DW, Ryoo SM, Lee YS, et al. An increase in initial shock index is associated with the requirement for massive transfusion in emergency department patients with primary postpartum hemorrhage. Shock. 2013;40: 101–105. doi: 10.1097/SHK.0b013e31829b1778 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref009] 9.Borovac-Pinheiro A, Ribeiro FM, Morais SS, Pacagnella RC. Shock index and heart rate standard reference values in the immediate postpartum period: A cohort study. PLoS One. 2019;14: e0217907. doi: 10.1371/journal.pone.0217907 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref010] 10.Perera R, Heneghan C. Making sense of diagnostic test likelihood ratios. ACP J Club. 2007;146: A8–9 . [PubMed] [Google Scholar]

[pone.0258619.ref011] 11.Nahm CH, Choi JW, Lee J. Delta neutrophil index in automated immature granulocyte counts for assessing disease severity of patients with sepsis. Ann Clin Lab Sci. 2008;38: 241–246 . [PubMed] [Google Scholar]

[pone.0258619.ref012] 12.Kong T, Park YS, Lee HS, Kim S, Lee JW, You JS, et al. The delta neutrophil index predicts development of multiple organ dysfunction syndrome and 30-day mortality in trauma patients admitted to an intensive care unit: a retrospective analysis. Sci Rep. 2018;8: 17515. doi: 10.1038/s41598-018-35796-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref013] 13.Kong T, In S, Park YS, Lee HS, Lee JW, You JS, et al. Usefulness of the Delta Neutrophil Index to Predict 30-Day Mortality in Patients with Upper Gastrointestinal Bleeding. Shock. 2017;48: 427–435. doi: 10.1097/SHK.0000000000000878 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref014] 14.Kong T, Kim TH, Park YS, Chung SP, Lee HS, Hong JH, et al. Usefulness of the delta neutrophil index to predict 30-day mortality in patients with ST segment elevation myocardial infarction. Sci Rep. 2017;7: 15718. doi: 10.1038/s41598-017-15878-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref015] 15.Kong T, Park YS, Lee HS, Kim S, Lee JW, Yu G, et al. Value of the Delta Neutrophil Index for Predicting 28-Day Mortality in Patients With Acute Pulmonary Embolism in the Emergency Department. Shock. 2018;49: 649–657. doi: 10.1097/SHK.0000000000001027 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref016] 16.Yune HY, Chung SP, Park YS, Chung HS, Lee HS, Lee JW, et al. Delta neutrophil index as a promising prognostic marker in out of hospital cardiac arrest. PLoS One. 2015;10: e0120677. doi: 10.1371/journal.pone.0120677 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref017] 17.Kong T, Park YS, Lee HS, Kim S, Han S, Eun CH, et al. A Delta Neutrophil Index for the Prediction of Contrast-Induced Nephropathy in Patients With St-Elevation Myocardial Infarction Followed By Percutaneous Coronary Intervention. Shock. 2018;49: 317–325. doi: 10.1097/SHK.0000000000000957 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref018] 18.Kim H, Kong T, Chung SP, Hong JH, Lee JW, Joo Y, et al. Usefulness of the Delta Neutrophil Index as a Promising Prognostic Marker of Acute Cholangitis in Emergency Departments. Shock. 2017;47: 303–312. doi: 10.1097/SHK.0000000000000722 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref019] 19.Kim JH, Park YS, Yoon CY, Lee HS, Kim S, Lee JW, et al. Delta Neutrophil Index for the Prediction of the Development of Sepsis-Induced Acute Kidney Injury in the Emergency Department. Shock. 2019;52: 414–422. doi: 10.1097/SHK.0000000000001299 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref020] 20.Kong T, Park YS, Lee HS, Kim S, Lee JW, You JS, et al. The Delta Neutrophil Index Predicts the Development of In-hospital Hypotension in Initially Stable Patients with Pyogenic Liver Abscess. Sci Rep. 2019;9: 12105. doi: 10.1038/s41598-019-48588-1 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref021] 21.Park JH, Byeon HJ, Lee KH, Lee JW, Kronbichler A, Eisenhut M, et al. Delta neutrophil index (DNI) as a novel diagnostic and prognostic marker of infection: a systematic review and meta-analysis. Inflamm Res. 2017;66: 863–870. doi: 10.1007/s00011-017-1066-y . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref022] 22.Manz MG, Boettcher S. Emergency granulopoiesis. Nat Rev Immunol. 2014;14: 302–314. doi: 10.1038/nri3660 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref023] 23.Cain DW, Snowden PB, Sempowski GD, Kelsoe G. Inflammation triggers emergency granulopoiesis through a density-dependent feedback mechanism. PLoS One. 2011;6: e19957. doi: 10.1371/journal.pone.0019957 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref024] 24.Leliefeld PH, Wessels CM, Leenen LP, Koenderman L, Pillay J. The role of neutrophils in immune dysfunction during severe inflammation. Crit Care. 2016;20: 73. doi: 10.1186/s13054-016-1250-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref025] 25.Alves-Filho JC, Spiller F, Cunha FQ. Neutrophil paralysis in sepsis. Shock. 2010;34 Suppl 1: 15–21. doi: 10.1097/SHK.0b013e3181e7e61b . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref026] 26.Livingston DH, Anjaria D, Wu J, Hauser CJ, Chang V, Deitch EA, et al. Bone marrow failure following severe injury in humans. Ann Surg. 2003;238: 748–753. doi: 10.1097/01.sla.0000094441.38807.09 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0258619.ref027] 27.Livingston DH, Gentile PS, Malangoni MA. Bone marrow failure after hemorrhagic shock. Circ Shock. 1990;30: 255–263 . [PubMed] [Google Scholar]

[pone.0258619.ref028] 28.Guérin E, Orabona M, Raquil MA, Giraudeau B, Bellier R, Gibot S, et al. Circulating immature granulocytes with T-cell killing functions predict sepsis deterioration*. Crit Care Med. 2014;42: 2007–2018. doi: 10.1097/CCM.0000000000000344 . [DOI] [PubMed] [Google Scholar]

[pone.0258619.ref029] 29.Figueiredo S, Taconet C, Harrois A, Hamada S, Gauss T, Raux M, et al. How useful are hemoglobin concentration and its variations to predict significant hemorrhage in the early phase of trauma? A multicentric cohort study. Ann Intensive Care. 2018;8: 76. doi: 10.1186/s13613-018-0420-8 . [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Delta neutrophil index and shock index can stratify risk for the requirement for massive transfusion in patients with primary postpartum hemorrhage in the emergency department

Taeyoung Kong

Hye Sun Lee

So Young Jeon

Je Sung You

Jong Wook Lee

Hyun Soo Chung

Sung Phil Chung

Roles

Abstract

Background

Methods

Results

Conclusion

Introduction

Methods

Study design and patients

Fig 1. Flow diagram of the enrollment of patients.

Data collection

Patients’ characteristics

Table 1. Comparison of clinical characteristics and clinical outcomes according to the requirement for massive transfusion in patients with primary postpartum hemorrhage.

DNI measurement

SI measurement

Clinical endpoints

Statistical analysis

Results

DNI and SI as predictors of the requirement for massive transfusion

Table 2. Factors associated with the requirement for massive transfusion in patients with primary postpartum hemorrhage in multivariate analysis.

Fig 2. Receiver operating characteristic curves of the delta neutrophil index, shock index, and the combination of the delta neutrophil index and shock index for the requirement for massive transfusion in patients with primary postpartum hemorrhage.

Table 3. Prognostic value of the shock index, delta neutrophil index, and combined delta neutrophil index with shock index for predicting the requirement for a massive transfusion in patients with primary postpartum hemorrhage.

Prognostic value of the combination of the DNI and SI for the requirement for massive transfusion

Discussion

Conclusions

Supporting information

Abbreviations

Data Availability

Funding Statement

References

Decision Letter 0

Wisit Cheungpasitporn

Roles

Author response to Decision Letter 0

Decision Letter 1

Wisit Cheungpasitporn

Roles

Author response to Decision Letter 1

Decision Letter 2

Wisit Cheungpasitporn

Roles

Acceptance letter

Wisit Cheungpasitporn

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases