New clinical prediction model for early recognition of sepsis in adult primary care patients: a prospective diagnostic cohort study of development and external validation

Feike J Loots; Marleen Smits; Rogier M Hopstaken; Kevin Jenniskens; Fleur H Schroeten; Ann van den Bruel; Alma C van de Pol; Jan Jelrik Oosterheert; Hjalmar Bouma; Paul Little; Michael Moore; Sanne van Delft; Douwe Rijpsma; Joris Holkenborg; Bas CT van Bussel; Ralph Laven; Dennis CJJ Bergmans; Jacobien J Hoogerwerf; Gideon HP Latten; Eefje GPM de Bont; Paul Giesen; Annemarie den Harder; Ron Kusters; Arthur RH van Zanten; Theo JM Verheij

doi:10.3399/BJGP.2021.0520

Published eLetters

eLetter submission guidelines

Submit a Response to This Article

Compose eLetter

Title *

Author Information

Contributors

First Name and Middle Initial *

First or given name, e.g. 'Peter'.

Last Name *

Your last, or family, name, e.g. 'MacMoody'.

Email Address *

Your email address, e.g. higgs-boson@gmail.com. PLEASE NOTE: your email address will be published.

Occupation *

Your role and/or occupation, e.g. 'Orthopedic Surgeon'.

Affiliation *

Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.

Statement of Competing Interests

Competing interests? *

Yes

Please describe the competing interests

CAPTCHA

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

Jump to comment:

Shaking chills may be better than rigors for sepsis prediction
Feike Loots

Published on: 07 June 2022
Shaking chills may be better than rigors for sepsis prediction.
Junki Mizumoto

Published on: 30 May 2022
Development and external validation of a new clinical prediction model for early recognition of sepsis
Feike Loots

Published on: 12 April 2022
Development and external validation of a new clinical prediction model for early recognition of sepsis
Diarmuid Quinlan

Published on: 06 April 2022

Published on: (7 June 2022)
Page navigation anchor for Shaking chills may be better than rigors for sepsis prediction
Shaking chills may be better than rigors for sepsis prediction
Feike Loots, MD, PhD candidate, University Medical Center Utrecht
We thank Junki Mizumoto for the interest in our study. We acknowledge that the study of Tokuda et al used a more detailed description of the severity of chills. The association between shivering and bacteremia is proven more convincingly by showing the relative risk increased in more severe categories of chills. However, our data showed no relation between rigors and the outcome sepsis. Rigors corresponds with shaking chills and it is very unlikely that we would have found an association, if we had used the same categorisation used in the study of Tokuda et al. Also, it is important to note that bacteremia and sepsis are not the same. Shaking chills or rigors might be useful for the decision to prescribe antibiotics, but our study results do not show evidence of added value in the decision to refer a patient to the hospital for (possible) sepsis.

Competing Interests: None declared.
Published on: (30 May 2022)
Page navigation anchor for Shaking chills may be better than rigors for sepsis prediction.
Shaking chills may be better than rigors for sepsis prediction.
Junki Mizumoto, Family physician, Department of Medical Education Studies, International Research Center for Medical Education, The University of Tokyo
I read the article by Loots FJ et al with great interest, and strongly appreciate the importance of the study.¹ The authors selected rigors (yes/no) as one of the candidate clinical predictors. The item of rigors was excluded in the simplified model because univariable and multivariable regression analyses showed no significant relationship between rigors and the diagnosis of sepsis.
The authors judged whether a patient developed rigors or not in a dichotomous manner. However, the degree of chills is important in predicting bacteremia. For example, shaking chills showed a specificity of 90.3% and a positive likelihood ratio (PLR) of 4.65 for bacteremia, while mild chills showed a specificity of 51.6% and PLR of 1.81 (see Tokuda et al.)²Considering a higher specificity, shaking chills may be a more desirable candidate than rigors only.

References

1. Loots FJ, Smits M, Hopstaken RM, et al. New clinical prediction model for early recognition of sepsis in adult primary care patients: a prospective diagnostic cohort study of development and external validation. Br J Gen Pract 2022;72(719):e437-e445. doi:10.3399/BJGP.2021.0520.

2. Tokuda Y, Miyasato H, Stein GH, Kishaba T. The degree of chills for risk of bacteremia in acute febrile illness. Am J Med 2005;118(12):1417. doi:10.1016/j.amjmed.2005.06.043.

Competing Interests: None declared.
Published on: (12 April 2022)
Page navigation anchor for Development and external validation of a new clinical prediction model for early recognition of sepsis
Development and external validation of a new clinical prediction model for early recognition of sepsis
Feike Loots, MD, PhD candidate, University Medical Center Utrecht
We thank Diarmuid Quinlan for the interest in our study, and he correctly points out that the patients included in the study have a substantially higher risk compared to the average adult patient assessed by a GP. We do not advocate to use the new derived model to screen for sepsis in all adult patients with a suspected infection presenting in primary care, as this would unnecessarily increase the workload for GPs and result in many false positive cases. In our view, a predictive model should be used only in cases where a GP perceives a clinical problem. Therefore, we decided to include only study subjects with relevant signs and/or symptoms and those in whom the GP suspected a serious infection. This however did not lead to a study population with “only the sickest patients” The proportion of patients meeting the Sepsis-3 criteria was high (42%), but this does not mean all patients were critically ill at the time of inclusion. The rate of referral to the hospital after the index consultation was 56% and 3% of the patients were admitted to the ICU.

We do not feel the phrasing “early recognition of sepsis” in the title is misleading. The population included in the study is a high-risk population for sepsis, but we included patients directly after the first contact with the out-of-hours GP cooperative and almost half of the patients were not referred to the hospital after inclusion. We expect the model to be valid for all patients suspec...
Show More

We thank Diarmuid Quinlan for the interest in our study, and he correctly points out that the patients included in the study have a substantially higher risk compared to the average adult patient assessed by a GP. We do not advocate to use the new derived model to screen for sepsis in all adult patients with a suspected infection presenting in primary care, as this would unnecessarily increase the workload for GPs and result in many false positive cases. In our view, a predictive model should be used only in cases where a GP perceives a clinical problem. Therefore, we decided to include only study subjects with relevant signs and/or symptoms and those in whom the GP suspected a serious infection. This however did not lead to a study population with “only the sickest patients” The proportion of patients meeting the Sepsis-3 criteria was high (42%), but this does not mean all patients were critically ill at the time of inclusion. The rate of referral to the hospital after the index consultation was 56% and 3% of the patients were admitted to the ICU.

We do not feel the phrasing “early recognition of sepsis” in the title is misleading. The population included in the study is a high-risk population for sepsis, but we included patients directly after the first contact with the out-of-hours GP cooperative and almost half of the patients were not referred to the hospital after inclusion. We expect the model to be valid for all patients suspected of sepsis or otherwise having signs of a serious infection in the primary care setting, but as we stated in the article, additional research is needed before widely adopting the model in practice.

Feike Loots, on behalf of the co-authors.

Show Less

Competing Interests: None declared.
Published on: (6 April 2022)
Page navigation anchor for Development and external validation of a new clinical prediction model for early recognition of sepsis
Development and external validation of a new clinical prediction model for early recognition of sepsis
Diarmuid Quinlan, GP, ICGP
The inclusion of criteria selected only the sickest of patients, with 42% ultimately being diagnosed with sepsis. The inclusion criteria were "Acutely ill adult (≥18 years) patients with fever, confusion, general deterioration or otherwise suspected severe infection were eligible for inclusion." The exclusion of respiratory rate as "less feasible for GPs to perform" merits further consideration of the wider implications.

The title of this article may be misleading: The presumptive diagnosis of sepsis in a confused, feverish adult with "suspected severe infection" is not an 'Early' diagnosis. The title might be amended to '...a new clinical prediction model for accurate recognition of sepsis in housebound seriously ill adults'. The applicability and validity of this scoring model in 'routine GP Out-of-Hours' is unproven.

Competing Interests: None declared.