Potential preanalytical error in arterial blood gases examined

The gold standard sample for blood gas analysis is arterial blood. Compared with collection of venous or capillary blood, collection of arterial blood by arterial puncture is a painful procedure that can induce feelings of anxiety. 

The notion that this pain and anxiety could be a source of preanalytical error in blood gas testing is born of the observation that both severe pain, such as that which occurs during labor, and acute anxiety (panic) attack cause increase in respiratory rate (hyperventilation) with reduction in pCO₂ and consequent increase in pH (i.e. acute respiratory alkalosis). 

But is the pain/anxiety experienced by patients during arterial puncture sufficient to cause hyperventilation and materially affect blood gas (pCO₂, pH) results? Although some clinicians are apparently willing to accept such a view when interpreting test results, the evidence base is severely lacking. 

This recently published study of 24 patients referred to a pulmonary function laboratory for blood gas testing, was designed to determine if pain/anxiety-induced hyperventilation occurs during arterial puncture. 

Investigators employed capnography during arterial puncture for this study. When attached via a nasal cannula, the capnograph allows continuous displayed monitoring of four parameters: end tidal carbon dioxide pressure (P_ETCO₂); oxygen saturation by pulse oximetry (SpO₂); respiratory rate (RR); and heart rate (HR). 

The capnograph was connected to each study participant prior to arterial puncture and the displayed parameters were noted at four time points: 15 mins before arterial puncture; during local anesthesia by lidocaine injection; during arterial puncture; and finally, 15 minutes after arterial puncture. 

All study participants completed a well-validated questionnaire to measure anxiety and rated the pain associated with their arterial puncture on a scale of 0 (no pain) to 10 (worst pain imaginable). 

Although there was a slight increase in mean respiratory rate of the 24 patients (19 ± 6 breaths per minute at baseline to 21 ± 6 breaths per minute during arterial puncture), mean P_ETCO₂ remained unchanged at 35 ± 5 mmHg from baseline through arterial puncture to 15 minutes post arterial puncture. The slight change in respiratory rate did not correlate with pain and anxiety scores. 

The authors conclude that arterial puncture is associated with small and brief increase in respiratory rate that is insufficient to affect P_ETCO₂ and therefore insufficient to affect blood gas results (pCO₂ and pH). A borderline or mildly elevated pCO₂ and pH should not be interpreted as due to the pain/anxiety of arterial puncture. 

In discussion of their study, the authors report that their results support those of the only other study to address this issue. There is currently no evidence to suggest that the pain/anxiety patients suffer during arterial puncture affects blood gas results.