The value of diaphragmatic ultrasound combined with procalcitonin, B-type natriuretic peptide, and lactate in assessing short-term prognosis of sepsis

Introduction

Sepsis, a life-threatening condition triggered by dysregulated host response to infection, frequently progresses to septic shock and multi-organ failure. In clinical practice, early assessment of disease severity and prognosis facilitates timely therapeutic intervention, thereby improving patient outcomes and reducing mortality. Current prognostic tools, including clinical scoring systems such as Sequential Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation II (APACHE II), have significant limitations. These tools primarily reflect systemic physiological derangements rather than dysfunction in specific organs. Diaphragmatic dysfunction, an underrecognized complication in sepsis, occurs in up to 64% of cases and is associated with a mortality rate of 37% (1). As the principal muscle of inspiration, the diaphragm accounts for 60–70% of tidal volume (2). Although transdiaphragmatic pressure measurement remains the gold standard for assessing diaphragmatic function, its invasive nature and technical demands limit clinical applicability (3).

The selection of procalcitonin (PCT), B-type natriuretic peptide (BNP), lactate (Lac), and diaphragmatic ultrasound parameters as core predictive indicators in this study is primarily based on the multi-system pathophysiological mechanisms underlying sepsis prognosis. Specifically, PCT serves as a specific marker reflecting the severity of systemic bacterial infection; elevated BNP levels indicate sepsis-associated myocardial suppression and cardiac dysfunction; lactate is a key indicator for assessing tissue hypoperfusion and hypoxic metabolism (4-8). However, these circulating biomarkers cannot directly evaluate the organ-specific function of respiratory failure, which is a common complication and cause of death in septic patients. Diaphragmatic ultrasound (by measuring excursion and thickening fraction) enables non-invasive, real-time quantification of diaphragmatic function, directly reflecting the reserve capacity of the respiratory pump (9). Therefore, the combined assessment of these indicators—which respectively reflect infection, circulation, metabolism, and respiratory function—holds promise for capturing the key physiological disturbances determining sepsis prognosis more comprehensively from multiple dimensions, overcoming the limitations of single-parameter approaches. Furthermore, all these indicators can be rapidly obtained at the bedside, demonstrating strong potential for clinical translation. This study aims to evaluate the individual and combined prognostic utility of diaphragmatic ultrasonography and serum biomarkers in septic patients, proposing an integrated assessment approach to optimize clinical decision-making. We present this article in accordance with the STARD reporting checklist (available at https://amj.amegroups.com/article/view/10.21037/amj-25-67/rc).

Methods

Study population

A retrospective analysis was conducted on septic patients meeting Sepsis-3 criteria admitted to Beijing Chaoyang Hospital from September 2021 to June 2023 randomly (10). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics board of Beijing Chaoyang Hospital, Capital Medical University (Beijing, China) (No. 2021-ke-704) and individual consent for this retrospective analysis was waived. Inclusion criteria comprised: (I) age ≥18 years; (II) not under invasive mechanical ventilation at enrollment; (III) sepsis diagnosis with biomarker and ultrasound data acquired within 24 hours of admission. Exclusion criteria included pre-existing neuromuscular disorders, thoracic deformities, recent neuromuscular blocker use, pregnancy, or impaired consciousness. Based on 28-day sepsis-specific mortality, the patients were categorized into a non-survivor group and a survivor group. During the 28-day period, a total of 7 patients were discharged or transferred to other facilities. Reasons included: stabilization of symptoms, need for specialized care, transfer to rehabilitation units, or patient preference. Telephone follow-up was performed for all patients who were discharged or transferred to guarantee the precision of outcome evaluation.

Data collection

This study was a single-center retrospective observational study. All data analyzed were obtained from routine clinical records. Diaphragmatic ultrasound assessment is an established early bedside examination protocol for patients with sepsis or those at risk of respiratory dysfunction in the intensive care unit (ICU) of our hospital. Therefore, this study systematically reviewed diaphragmatic ultrasound examination results and relevant biomarker data—completed according to clinical routine within 24 hours of admission—for all eligible septic patients during the study period. Clinical variables included age, gender, APACHE II score, comorbidities, and serum PCT, BNP, Lac levels. Patients’ respiratory support modalities including non-invasive ventilation (NIV) and high-flow nasal cannula (HFNC) oxygen therapy have been thoroughly documented. Prognostic indicators: 28-day survival status after admission was used as the primary endpoint.

Ultrasound equipment and methods

Diaphragmatic ultrasound was performed within 24 hours after diagnosis using a Mindray I9 bedside ultrasound system with a 1–5 MHz convex array probe and an 8–10 MHz linear array probe. Patients were placed in a supine position, and the right diaphragm was measured.

(I) Diaphragm excursion (DE): the convex array probe was placed obliquely at the midpoint between the right anterior axillary line and midclavicular line in the 8^th–9^th intercostal spaces. M-mode ultrasound was used to record diaphragmatic movement during quiet breathing. DE was DE was measured as the vertical distance between the peak and trough of the motion curve. Measurements were repeated three times and averaged. (II) Diaphragmatic thickening fraction (DTF): the linear array probe was placed transversely at the right midclavicular line in the 8th–9th intercostal space. Diaphragmatic thickness was defined as the hypoechoic muscle layer between the pleural and peritoneal layers. Diaphragm thickness at end-inspiration (DTei) and end-expiration (DTee) were measured three times each and averaged. DTF was calculated as:

DTF=(DTei − DTee) / DTee × 100%.

Measurement methods are illustrated in Figures 1,2.

Figure 1 Measurement of the diaphragmatic thickness at the end-inspiration (A) and end-expiration (B). DTee, diaphragm thickness at end-expiration; DTei, diaphragm thickness at end-inspiration.

Figure 2 Measurement of DE. DE, diaphragm excursion.

Data quality control

To ensure the accuracy of diaphragmatic measurements, ultrasound images were deemed inadequate for analysis if the pleural and peritoneal lines could not be clearly visualized to define the diaphragmatic thickness. All patients for whom only such inadequate images were available were excluded from the study. Furthermore, any participant with missing data for key clinical variables or outcomes was excluded from the final analysis via complete-case analysis.

All ultrasound image acquisition and measurements were performed by two certified sonographers, each with over 5 years of clinical experience in scanning and interpretation. To minimize bias, the personnel conducting the image analysis were blinded to all clinical information (including patient group assignment, treatment regimens, and laboratory parameters) and the 28-day outcomes. Prior to the study, both operators completed a unified training program on the study-specific measurement protocol, which included standardized image acquisition criteria and blinded measurement procedures, to ensure consistency in operation and assessment.

Statistical analysis

Statistical analyses were performed using SPSS 26.0. Normally distributed continuous variables are expressed as mean ± standard deviation and compared using the independent samples t-test, while non-normally distributed continuous variables are expressed as median (interquartile range) and compared using the Mann-Whitney U test. Categorical variables are presented as frequencies and compared using the Chi-squared test. Binary logistic regression and receiver operating characteristic (ROC) analyses identified independent predictors and evaluated prognostic performance.

Results

Comparison of general data

A total of 74 patients with sepsis were included in this study. Based on 28-day mortality, the patients were categorized into a non-survivor group (n=29) and a survivor group (n=45). No statistically significant differences were observed between the two groups regarding age, gender, underlying diseases, APACHE II scores or respiratory support (all P>0.05), indicating that the baseline characteristics were well-balanced and comparable. See Table 1. Figure 3 shows the flowchart.

Figure 3 Flowchart.

Comparison of laboratory parameters within 24 hours of admission

Serum levels of PCT, BNP, and Lac measured within 24 hours after admission were significantly higher in the non-survivor group compared to the survivor group (all P<0.05). Refer to Table 2 for details.

Comparison of diaphragmatic ultrasound parameters between the two group

DTei, DTee, DTF, and DE were all significantly reduced in the non-survivor group relative to the survivor group (all P<0.05). The results are summarized in Table 3.

Binary logistic regression analysis of prognostic factors

The univariate analysis revealed that serum levels of PCT, BNP, Lac, DTF, and DE were significantly associated with the 28-day mortality rate in sepsis patients (all P<0.05). However, following adjustment by multivariable logistic regression, only DTF and DE remained independent predictors of prognosis (both P<0.05). See Table 4.

ROC curve analysis

The combined use of these indicators exhibited the highest predictive performance, with an area under the curve (AUC) of 0.890 [95% confidence interval (CI): 0.818–0.961] a sensitivity of 96.6% and a specificity of 57.80% (P<0.001). Detailed values are presented in Table 5.

Discussion

This study systematically evaluated and validated the combined value of diaphragmatic ultrasonographic functional parameters and serum biomarkers for predicting short-term prognosis in patients with sepsis. The key findings are as follows: first, the combined use of five indicators (DE, DTF, PCT, BNP and Lac) yielded the highest predictive performance for 28-day mortality, with AUC of 0.890 and a sensitivity of 96.6%, significantly outperforming traditional scoring systems such as SOFA or APACHE II reported in the literature (11-13). Second, in the early admission period (within 24 hours), non-survivors showed significantly lower diaphragmatic contractile function indices (including DTei, DTee, DTF, and DE) and significantly higher serum inflammatory and stress markers (PCT, BNP, and Lac). Third, ROC analysis demonstrated high specificity for traditional biomarkers (PCT 97.8%, BNP 80.0%, Lac 93.3%). While univariate analysis revealed significant associations between these indicators and mortality, multivariate adjustment identified only DTF and DE as independent predictors of 28-day mortality, whereas PCT, BNP, and Lac required combination with diaphragmatic function parameters to further enhance predictive efficacy. These findings suggest that the prognosis of sepsis patients is influenced by the synergistic effects of systemic inflammatory response, cardiac function status, diaphragmatic structural and functional integrity, and tissue perfusion levels. The five-parameter combined model integrates multidimensional pathophysiological information, highlighting the complementary value of combined assessment.

The strength of this study lies in the integration of objective, quantifiable diaphragmatic ultrasonographic parameters with readily available serum biomarkers to establish a simple, bedside-applicable early warning model. Its high sensitivity is particularly suitable for early risk screening in critically ill patients. However, several limitations should be acknowledged: first, the single-center retrospective design and modest sample size may limit the generalizability of the findings, and insufficient statistical power could lead to false negatives. Second, all parameters were measured only at a single time point upon admission, precluding analysis of their dynamic changes in relation to prognosis. Finally, although widely accepted ultrasonographic surrogate measures were used, the study did not employ transdiaphragmatic pressure, the gold standard, for synchronous validation.

Previous studies have largely focused on the combination of indicators of a single type. A study involving septic patients receiving mechanical ventilation demonstrated DE and DTF achieved AUC values of 0.85 and 0.97, respectively, in predicting successful weaning; however, it did not evaluate mortality prediction or incorporate traditional biomarkers (14). Several other studies have likewise confirmed that improvement in BNP levels during ICU hospitalization is closely associated with increased survival rates, and the magnitude of change between baseline BNP values and measurements taken at 72 hours exhibits a statistically significant correlation with 28-day mortality (15,16). In contrast, the five-parameter combined model in this study achieved an AUC of 0.890 and a sensitivity of 96.6%, significantly outperforming the predictive performance of previous single-type indicator combinations. Consistent with our findings, existing research has confirmed that sepsis-induced diaphragmatic dysfunction is closely associated with increased mortality, and while PCT, BNP, and Lac individually exhibit high specificity, their sensitivity is insufficient. By combining “diaphragmatic function indicators and traditional biomarkers”, this study achieves complementary “functional status and pathophysiological indicators”: diaphragmatic function parameters reflect the patient’s core organ reserve capacity, while traditional biomarkers reveal specific types of pathophysiological disturbances. Their combination enables more accurate identification of high-risk patients.

From a pathophysiological perspective, the core of sepsis is a cascade reaction involving “cytokine storm → multi-organ dysfunction → tissue hypoperfusion → metabolic disturbances → multi-organ failure”. As the primary muscle responsible for respiratory load, the functional integrity of the diaphragm directly determines alveolar ventilation efficiency and systemic oxygenation. During sepsis, pro-inflammatory factors such as endotoxin, TNF-α, and IL-6 bind to Toll-like receptor 4 on diaphragmatic cells, activating the NF-κB signaling pathway (17). This leads to cellular edema and necrosis on one hand and inhibits calcium signaling pathways on the other, resulting in weakened interactions between diaphragmatic contractile proteins and reduced contraction strength. Concurrently, the hypermetabolic and catabolic state induced by sepsis impairs diaphragmatic protein synthesis and accelerates muscle breakdown, further diminishing contractile function, manifested as reduced DTF (18). Decreased diaphragmatic contraction strength directly limits movement amplitude. Moreover, reduced lung compliance due to sepsis increases respiratory resistance, leading to chronic overload and muscle fatigue. Additionally, sepsis can damage the phrenic nerve, disrupting coordination of diaphragmatic contraction, collectively contributing to significantly reduced DE.

At the biomarker level, endotoxin during sepsis induces systemic synthesis and release of PCT, whose levels positively correlate with the intensity of the inflammatory response, helping identify high-risk patients with “uncontrolled inflammation”. However, PCT alone has limited sensitivity and may be influenced by early antibiotic therapy; combining it with diaphragmatic function parameters can reduce misjudgments due to therapeutic interventions. Sepsis-induced acute cardiac dysfunction and pulmonary congestion increase ventricular wall tension, stimulating BNP release. This marker aids in identifying prognosis risks related to “impaired cardiac function”, and its combination with diaphragmatic function parameters can guide intervention strategies (19). Diaphragmatic dysfunction leads to inadequate alveolar ventilation and microcirculatory impairment, causing tissue hypoxia and activating anaerobic glycolysis, resulting in lactate accumulation. Lactate levels directly reflect the severity of tissue hypoxia, and combining them with diaphragmatic function parameters validates the impact of diaphragmatic dysfunction on systemic metabolism. Thus, diaphragmatic function indicators offer high sensitivity for identifying potential high-risk patients, while traditional biomarkers provide high specificity to reduce false positives. Their combination endows the model with both high sensitivity and high specificity.

This study suggests that incorporating bedside diaphragmatic ultrasound into the early assessment protocol for sepsis patients can effectively identify high-risk individuals and enable earlier intervention. Future efforts should include: first, conducting prospective, multicenter, large-scale cohort studies to externally validate and optimize this combined prediction model. Second, performing longitudinal studies to investigate whether dynamic changes in DTF and DE can more precisely guide treatment response and prognosis evaluation. Third, exploring whether early intervention strategies based on this model can improve clinical outcomes in high-risk patients. Fourth, promoting the development of standardized protocols for diaphragmatic ultrasound image acquisition and measurement to enhance reproducibility and clinical feasibility.

Conclusions

This study demonstrates that the combination of diaphragmatic ultrasonographic parameters with serum biomarkers (procalcitonin, brain natriuretic peptide, and lactate) provides significant predictive value for the short-term prognosis (28-day mortality) of sepsis patients. These findings suggest that this multimodal assessment method holds potential as a clinically useful tool for early prognosis. Future research should incorporate broader clinical endpoints (such as in-hospital mortality and 60-/90-day mortality) and conduct long-term follow-up to clarify the predictive capability of this combined indicator for patients’ long-term survival and functional recovery, thereby enabling a comprehensive evaluation of its ultimate potential for clinical translation.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://amj.amegroups.com/article/view/10.21037/amj-25-67/rc

Data Sharing Statement: Available at https://amj.amegroups.com/article/view/10.21037/amj-25-67/dss

Peer Review File: Available at https://amj.amegroups.com/article/view/10.21037/amj-25-67/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://amj.amegroups.com/article/view/10.21037/amj-25-67/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics board of Beijing Chaoyang Hospital, Capital Medical University (Beijing, China) (No. 2021-ke-704) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Maloney PJ. Sepsis and septic shock. Emerg Med Clin North Am 2013;31:583-600. [Crossref] [PubMed]
2. Kocjan J, Adamek M, Gzik-Zroska B, Czyżewski D, Rydel M. Network of breathing. Multifunctional role of the diaphragm: a review. Adv Respir Med 2017;85:224-32.
3. Tullo G, Candelli M, Gasparrini I, et al. Ultrasound in Sepsis and Septic Shock-From Diagnosis to Treatment. J Clin Med 2023;12:1185. [Crossref] [PubMed]
4. Boon AJ, Sekiguchi H, Harper CJ, et al. Sensitivity and specificity of diagnostic ultrasound in the diagnosis of phrenic neuropathy. Neurology 2014;83:1264-70. [Crossref] [PubMed]
5. Baranowsky A, Appelt J, Kleber C, et al. Procalcitonin Exerts a Mediator Role in Septic Shock Through the Calcitonin Gene-Related Peptide Receptor. Crit Care Med 2021;49:e41-52. [Crossref] [PubMed]
6. Puchongmart C, Nakornchai T, Leethotsarat K, et al. The Incidence of Diaphragmatic Dysfunction in Patients Presenting With Dyspnea in the Emergency Department. J Ultrasound Med 2023;42:1557-66. [Crossref] [PubMed]
7. Nascimento TS, de Queiroz RS, Ramos ACC, et al. Ultrasound Protocols to Assess Skeletal and Diaphragmatic Muscle in People Who Are Critically Ill: A Systematic Review. Ultrasound Med Biol 2021;47:3041-67. [Crossref] [PubMed]
8. Valverde Montoro D, García Soler P, Hernández Yuste A, et al. Ultrasound assessment of ventilator-induced diaphragmatic dysfunction in mechanically ventilated pediatric patients. Paediatr Respir Rev 2021;40:58-64. [Crossref] [PubMed]
9. Boussuges A, Gole Y, Blanc P. Diaphragmatic motion studied by m-mode ultrasonography: methods, reproducibility, and normal values. Chest 2009;135:391-400. [Crossref] [PubMed]
10. Shankar-Hari M, Phillips GS, Levy ML, et al. Developing a New Definition and Assessing New Clinical Criteria for Septic Shock: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016;315:775-87. [Crossref] [PubMed]
11. Mak MHW, Low JK, Junnarkar SP, et al. A prospective validation of Sepsis-3 guidelines in acute hepatobiliary sepsis: qSOFA lacks sensitivity and SIRS criteria lacks specificity (Cohort Study). Int J Surg 2019;72:71-7. [Crossref] [PubMed]
12. Xia Y, Zou L, Li D, et al. The ability of an improved qSOFA score to predict acute sepsis severity and prognosis among adult patients. Medicine (Baltimore) 2020;99:e18942. [Crossref] [PubMed]
13. VijayGanapathy S. Validation of APACHE II scoring system at 24 hours after admission as a prognostic tool in urosepsis: A prospective observational study. Investig Clin Urol 2017;58:453-9. [Crossref] [PubMed]
14. Eltrabili HH, Hasanin AM, Soliman MS, et al. Evaluation of Diaphragmatic Ultrasound Indices as Predictors of Successful Liberation From Mechanical Ventilation in Subjects With Abdominal Sepsis. Respir Care 2019;64:564-9. [Crossref] [PubMed]
15. Klouche K, Pommet S, Amigues L, et al. Plasma brain natriuretic peptide and troponin levels in severe sepsis and septic shock: relationships with systolic myocardial dysfunction and intensive care unit mortality. J Intensive Care Med 2014;29:229-37. [Crossref] [PubMed]
16. Guaricci AI, Santoro F, Paoletti Perini A, et al. Correlations between NT-proBNP, outcome and haemodynamics in patients with septic shock. Acta Cardiol 2015;70:545-52. [Crossref] [PubMed]
17. Jiang J, Chen Q, Chen X, et al. Magnesium sulfate ameliorates sepsis-induced diaphragm dysfunction in rats via inhibiting HMGB1/TLR4/NF-κB pathway. Neuroreport 2020;31:902-8. [Crossref] [PubMed]
18. Jiao GY, Liu YJ, Yang KY, et al. Ultrasound study for quantitative assessment of diaphragm dysfunction in patients with sepsis. Zhonghua Jie He He Hu Xi Za Zhi 2025;48:116-22. [Crossref] [PubMed]
19. Wacker C, Prkno A, Brunkhorst FM, et al. Procalcitonin as a diagnostic marker for sepsis: a systematic review and meta-analysis. Lancet Infect Dis 2013;13:426-35. [Crossref] [PubMed]