Background Capillary blood gas (CBG) analysis may provide a less invasive alternative to arterial blood gas (ABG) sampling for selected measurements during recovery from critical illness. Its reliability and suitability for ICU de-escalation require evaluation. Aim The primary aim was to assess CBG agreement with comparator methods, principally ABG, in adults and identify clinically reliable parameters. Secondary aims were to examine factors affecting performance and sampling feasibility, applicability to ICU recovery, and potential clinical, organisational and economic implications for de-escalating invasive monitoring. Methods: PubMed, Embase (Ovid), Scopus and supplementary sources were searched for English-language publications dated 1 January 2000–30 September 2025. Adult hospital-based studies comparing capillary measurements with arterial, venous, laboratory, transcutaneous or pulse-oximetry methods were eligible. Paediatric, pre-hospital, laboratory-only, glucose-only and healthy-volunteer experimental studies were excluded. Parameters included pH, pCO₂, pO₂, bicarbonate, base excess, lactate, oxygen saturation, haemoglobin, haematocrit, electrolytes and creatinine. Reporting completeness was assessed using modified STROBE, STARD-based and PRISMA 2020 checklists. Methodological appraisal used adapted QAREL for numerical comparisons; risk of bias was assessed using QUADAS-3 for diagnostic results and ROBIS for the previous systematic review and meta-analysis. Findings were synthesised narratively by parameter, method and population; no new meta-analysis was performed. Results Across 27 primary studies, the reported study cohorts totalled 3,011 participants. The number contributing usable capillary comparisons varied by study and measurement. One previously published systematic review and meta-analysis provided secondary context; its constituent studies and participants were not added to the primary-study totals. Agreement was most consistent for pH and pCO₂. Bicarbonate and several additional biochemical and haematological parameters met study-specific criteria in selected populations. Conventional capillary pO₂ did not support arterial substitution, and oxygen-saturation replacement was not established. Lactate findings were mixed. Performance and sampling feasibility varied with technique, arterialisation, perfusion and clinical context. Reporting completeness varied: modified STROBE scores ranged from 41.9% to 94.8%, STARD-based scores from 60.9% to 83.3%, and the secondary meta-analysis had a PRISMA completeness score of 76.2%. Of 297 QAREL judgments, 204 were unclear because of insufficient information. Twenty-one of 22 diagnostic results and the secondary meta-analysis were at high risk of bias. No primary study directly evaluated a CBG-based ICU de-escalation pathway. Limitations Heterogeneity and incomplete reporting limited interpretation. Study selection, extraction and appraisal were undertaken by the author without independent duplicate review. No formal certainty grading was applied; confidence was assessed narratively through methodological appraisal, risk of bias, consistency, precision and ICU applicability. Conclusions CBG appears most promising for assessment of acid–base status and ventilation in selected stable adult patients, particularly through pH and pCO₂ measurement, but it cannot be considered a universal substitute for ABG. Its limitations are especially important when accurate assessment of oxygenation is required. The available evidence supports further prospective evaluation specifically in stable or recovering ICU patients to determine whether CBG can be incorporated safely into structured de-escalation pathways, reduce unnecessary arterial sampling or invasive monitoring, and generate meaningful clinical, organisational and economic benefits.
Background Capillary blood gas (CBG) analysis may provide a less invasive alternative to arterial blood gas (ABG) sampling for selected measurements during recovery from critical illness. Its reliability and suitability for ICU de-escalation require evaluation. Aim The primary aim was to assess CBG agreement with comparator methods, principally ABG, in adults and identify clinically reliable parameters. Secondary aims were to examine factors affecting performance and sampling feasibility, applicability to ICU recovery, and potential clinical, organisational and economic implications for de-escalating invasive monitoring. Methods: PubMed, Embase (Ovid), Scopus and supplementary sources were searched for English-language publications dated 1 January 2000–30 September 2025. Adult hospital-based studies comparing capillary measurements with arterial, venous, laboratory, transcutaneous or pulse-oximetry methods were eligible. Paediatric, pre-hospital, laboratory-only, glucose-only and healthy-volunteer experimental studies were excluded. Parameters included pH, pCO₂, pO₂, bicarbonate, base excess, lactate, oxygen saturation, haemoglobin, haematocrit, electrolytes and creatinine. Reporting completeness was assessed using modified STROBE, STARD-based and PRISMA 2020 checklists. Methodological appraisal used adapted QAREL for numerical comparisons; risk of bias was assessed using QUADAS-3 for diagnostic results and ROBIS for the previous systematic review and meta-analysis. Findings were synthesised narratively by parameter, method and population; no new meta-analysis was performed. Results Across 27 primary studies, the reported study cohorts totalled 3,011 participants. The number contributing usable capillary comparisons varied by study and measurement. One previously published systematic review and meta-analysis provided secondary context; its constituent studies and participants were not added to the primary-study totals. Agreement was most consistent for pH and pCO₂. Bicarbonate and several additional biochemical and haematological parameters met study-specific criteria in selected populations. Conventional capillary pO₂ did not support arterial substitution, and oxygen-saturation replacement was not established. Lactate findings were mixed. Performance and sampling feasibility varied with technique, arterialisation, perfusion and clinical context. Reporting completeness varied: modified STROBE scores ranged from 41.9% to 94.8%, STARD-based scores from 60.9% to 83.3%, and the secondary meta-analysis had a PRISMA completeness score of 76.2%. Of 297 QAREL judgments, 204 were unclear because of insufficient information. Twenty-one of 22 diagnostic results and the secondary meta-analysis were at high risk of bias. No primary study directly evaluated a CBG-based ICU de-escalation pathway. Limitations Heterogeneity and incomplete reporting limited interpretation. Study selection, extraction and appraisal were undertaken by the author without independent duplicate review. No formal certainty grading was applied; confidence was assessed narratively through methodological appraisal, risk of bias, consistency, precision and ICU applicability. Conclusions CBG appears most promising for assessment of acid–base status and ventilation in selected stable adult patients, particularly through pH and pCO₂ measurement, but it cannot be considered a universal substitute for ABG. Its limitations are especially important when accurate assessment of oxygenation is required. The available evidence supports further prospective evaluation specifically in stable or recovering ICU patients to determine whether CBG can be incorporated safely into structured de-escalation pathways, reduce unnecessary arterial sampling or invasive monitoring, and generate meaningful clinical, organisational and economic benefits.
Capillary Blood Gas Analysis in Stable Adult ICU Patients: A Systematic Review and Critical Appraisal of Clinical, Organisational and Economic Implications for De-escalation of Care
CITTON, ELIA
2024/2025
Abstract
Background Capillary blood gas (CBG) analysis may provide a less invasive alternative to arterial blood gas (ABG) sampling for selected measurements during recovery from critical illness. Its reliability and suitability for ICU de-escalation require evaluation. Aim The primary aim was to assess CBG agreement with comparator methods, principally ABG, in adults and identify clinically reliable parameters. Secondary aims were to examine factors affecting performance and sampling feasibility, applicability to ICU recovery, and potential clinical, organisational and economic implications for de-escalating invasive monitoring. Methods: PubMed, Embase (Ovid), Scopus and supplementary sources were searched for English-language publications dated 1 January 2000–30 September 2025. Adult hospital-based studies comparing capillary measurements with arterial, venous, laboratory, transcutaneous or pulse-oximetry methods were eligible. Paediatric, pre-hospital, laboratory-only, glucose-only and healthy-volunteer experimental studies were excluded. Parameters included pH, pCO₂, pO₂, bicarbonate, base excess, lactate, oxygen saturation, haemoglobin, haematocrit, electrolytes and creatinine. Reporting completeness was assessed using modified STROBE, STARD-based and PRISMA 2020 checklists. Methodological appraisal used adapted QAREL for numerical comparisons; risk of bias was assessed using QUADAS-3 for diagnostic results and ROBIS for the previous systematic review and meta-analysis. Findings were synthesised narratively by parameter, method and population; no new meta-analysis was performed. Results Across 27 primary studies, the reported study cohorts totalled 3,011 participants. The number contributing usable capillary comparisons varied by study and measurement. One previously published systematic review and meta-analysis provided secondary context; its constituent studies and participants were not added to the primary-study totals. Agreement was most consistent for pH and pCO₂. Bicarbonate and several additional biochemical and haematological parameters met study-specific criteria in selected populations. Conventional capillary pO₂ did not support arterial substitution, and oxygen-saturation replacement was not established. Lactate findings were mixed. Performance and sampling feasibility varied with technique, arterialisation, perfusion and clinical context. Reporting completeness varied: modified STROBE scores ranged from 41.9% to 94.8%, STARD-based scores from 60.9% to 83.3%, and the secondary meta-analysis had a PRISMA completeness score of 76.2%. Of 297 QAREL judgments, 204 were unclear because of insufficient information. Twenty-one of 22 diagnostic results and the secondary meta-analysis were at high risk of bias. No primary study directly evaluated a CBG-based ICU de-escalation pathway. Limitations Heterogeneity and incomplete reporting limited interpretation. Study selection, extraction and appraisal were undertaken by the author without independent duplicate review. No formal certainty grading was applied; confidence was assessed narratively through methodological appraisal, risk of bias, consistency, precision and ICU applicability. Conclusions CBG appears most promising for assessment of acid–base status and ventilation in selected stable adult patients, particularly through pH and pCO₂ measurement, but it cannot be considered a universal substitute for ABG. Its limitations are especially important when accurate assessment of oxygenation is required. The available evidence supports further prospective evaluation specifically in stable or recovering ICU patients to determine whether CBG can be incorporated safely into structured de-escalation pathways, reduce unnecessary arterial sampling or invasive monitoring, and generate meaningful clinical, organisational and economic benefits.| File | Dimensione | Formato | |
|---|---|---|---|
|
CITTON_ELIA.pdf
Accesso riservato
Dimensione
2.1 MB
Formato
Adobe PDF
|
2.1 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/120776