Stability and long-term durability of raman spectroscopy

Abstract

Objective. The purpose of this study was to assess in the clinical setting the reliability and long-term stability of Raman spectroscopy as implemented in the RASCAL multiple gas analyzer, and to describe/analyze failure modes that manifest in regular use.Methods. Twenty-one RASCAL analyzers were tested for accuracy and precision. Without any prior external calibration or alignment, all gas analyzers were systematically tested over a consecutive 36-hour period with standard gas mixtures. Data were analyzed by evaluating the difference between the predicted value and observed value (bias residual) as reported by each RASCAL. All service data (29 months) also were analyzed for information on durability and failure modes.Results. The RASCAL exhibited a significant tendency to overread high concentrations of agent (isoflurane/enflurane); 4 of 16 instruments misread an agent by more than ±0.30%. Four of 16 instruments could not properly identify volatile agents in low concentrations (0.31 vol%). Inventory records show that water contamination led to the replacement of gas sample sets an average of 1.5 ± 1.2 times per case over the 29-month period. Although many instruments had not been externally calibrated for over 63 days, linearity proved acceptable for CO₂, O₂, N₂O, and N₂. A rationale for instrument behavior and major failure modes, based on the instrument design, was developed.Conclusions. The manufacturer’s suggested calibration intervals (30 days) were found to be more than adequate for reliable gas analysis using Raman spectroscopy. Without the benefit of frequent calibrations and as time passes, volatile agent quantification can be expected to drift significantly upward.