Diagnostic accuracy of capnography in verification of nasogastric tube placement: A prospective multicentre study

Technical Summary: Diagnostic Accuracy of Capnography in Nasogastric Tube Placement Verification

Problem Statement
Inadvertent insertion of nasogastric tubes (NGTs) into the airway is a critical iatrogenic complication that can lead to pulmonary aspiration and death. While radiographic confirmation is the established gold standard, it is not always immediately available, particularly in general wards, emergency departments, and community care settings. Current bedside alternatives, such as pH testing of aspirates, are frequently limited by the inability to obtain fluid (occurring in nearly half of cases in this study) or misleading pH readings. Previous systematic reviews suggested that end-tidal carbon dioxide (ETCO₂) detection via capnography could offer diagnostic accuracy comparable to radiography in critically ill adults, but robust evidence regarding its performance in general ward and emergency care populations, as well as its cost-effectiveness compared to radiography alone, remained limited.

Methodology
This study was a multicentre, prospective, observational diagnostic trial conducted across 21 public hospitals in Hong Kong, involving 39 study sites (medical/geriatric wards, ICUs, and emergency departments).

• Participants: 508 patients aged ≥18 years requiring NGT insertion for assessment, nutrition, drainage, or medication. Exclusion criteria included patients undergoing lifesaving interventions, those on invasive mechanical ventilation with high oxygen fractions or PEEP ≥10 cmH₂O, and those exhibiting signs of respiratory distress during insertion.
• Index Test: Capnography was performed using a bedside sidestream capnography device (Medtronic Capnostream 35) connected to the distal end of the NGT. ETCO₂ values were recorded after a minimum of six breaths or one minute. Three diagnostic thresholds were evaluated: >5 mmHg, >10 mmHg, and >15 mmHg. A positive test was defined by the presence of a normal CO₂ waveform and a positive ETCO₂ reading.
• Reference Standard: Radiography (chest and/or abdominal X-ray) interpreted by a physician or radiologist served as the reference standard for confirming tube placement.
• Blinding and Protocol: To minimize bias, nurses performing the capnography test were blinded to the results when ward nurses performed pH testing. Radiograph interpreters were blinded to the index test results. All participants underwent radiography regardless of the capnography result to prevent verification bias.
• Analysis: Diagnostic performance metrics (sensitivity, specificity, PPV, NPV, likelihood ratios) were calculated for each threshold. A cost analysis compared the costs of using radiography alone versus a strategy of capnography followed by radiography only for positive capnography results. Cost savings were estimated using a bias-corrected and accelerated bootstrapping method with 20,000 replications.

Key Results

• Prevalence: Among 508 participants, radiography confirmed correct gastric placement in 452 cases (89.0%). Only one case (0.2%) of inadvertent airway placement was identified.
• Diagnostic Accuracy: The single case of airway placement was correctly identified by capnography across all three thresholds (ETCO₂ = 32 mmHg). Consequently, the study reported 100% sensitivity and 100% negative predictive value (NPV) for all thresholds.
  • Specificity: Varied by threshold: 95.7% (>5 mmHg), 96.8% (>10 mmHg), and 97.6% (>15 mmHg).
  • Likelihood Ratios: Positive likelihood ratios ranged from 23.1 to 42.3, while negative likelihood ratios were 0 for all thresholds.
  • False Positives: Approximately 4.3% of cases showed ETCO₂ >5 mmHg, but waveform analysis revealed these were transient or non-square patterns inconsistent with pulmonary placement, often attributable to residual gastric gas or recent carbonated intake.
• Cost and Time:
  • The average cost per patient for radiography alone was HK$405.9.
  • Using capnography as a screening tool reduced the average cost to HK$195.6, HK$191.3, and HK$188.3 for the >5, >10, and >15 mmHg thresholds, respectively.
  • This resulted in estimated cost savings of HK$210.3, HK$214.5, and HK$217.5 per patient, respectively.
  • The average waiting time from the index test to radiographic confirmation was 148.6 minutes.

Significance and Claims
The authors claim this study provides the first robust evidence on the diagnostic accuracy of capnography for verifying NGT placement across a diverse population including critically ill patients, general ward inpatients, and emergency department admissions.

• Clinical Utility: The study concludes that capnography is a valuable first-line tool for excluding airway placement of NGTs, particularly when used alongside pH testing of aspirates. Its high sensitivity and NPV ensure that airway misplacements are not missed.
• Operational Efficiency: The simplicity of the measurement and the ability to rapidly identify airway placement (compared to the ~150-minute wait for X-ray) support prompt reinsertion or repositioning of misplaced tubes.
• Economic Impact: The study demonstrates significant cost savings when capnography is used to triage patients, reducing the reliance on immediate radiography for all insertions.
• Safety: No adverse events were reported. The authors emphasize that while capnography effectively rules out airway placement, it does not confirm correct gastric positioning; therefore, it is an adjunct to, not a replacement for, radiographic confirmation in cases of uncertainty or positive capnography results.

The paper maintains a modest tone regarding the low prevalence of airway misplacement (0.2%) in the cohort, acknowledging that while only one case was found, the study's prospective design and large sample size across 21 hospitals provide strong statistical precision for the reported sensitivity and specificity.