Untersuchungen zum Einsatz der Pulsoxymetrie bei prilocaininduzierter Methämoglobinämie

Zusammenfassung

In einer prospektiven, randomisierten Doppelblindstudie wurde die Beeinflussung der Meßgenauigkeit dreier verschiedener Pulsoxymeter durch Methämoglobin untersucht. 171 Patienten, die sich einem handchirurgischen Eingriff in Plexusanästhesie unterzogen, erhielten während des gesamten Untersuchungszeitraums 6 l O ₂ über eine Hudson-Maske. Bei 41 Patienten wurde die Plexusblockade mit Lidocain und bei 130 Patienten mit Prilocain durchgeführt. Vor Anlage der Plexusanästhesie sowie 15, 30, 60 und 120 min danach wurden folgende Pulsoxymeter angelegt: 1. Ohmeda BIOX 3700e, 2. Critikon Oxyshuttle, 3. Nellcor Pulsoxymeter N180. Gleichzeitig wurde eine periphervenöse Blutprobe auf Dyshämoglobine untersucht. Mit der Lidocaingruppe konnte gezeigt werden, daß es mit der beschriebenen Methodik möglich ist hyperoxische Bedingungen aufrechtzuerhalten. In der Prilocaingruppe ergaben sich erhebliche psO ₂ -Abfälle, die sich zwischen den drei Pulsoxymetern signifikant unterschieden. Ein Zusammenhang zwischen Met-Hb und psO ₂ -Abfall für alle Pulsoxymeter gleich konnte nicht gefunden werden, sondern müßte individuell wie folgt definiert werden: 1. Ohmeda BIOX 3700e: Met-Hb= (101-psO ₂ )·0,6 (r=0,94), 2. Critikon Oxyshuttle: Met-Hb= (101-psO ₂ )·0,7 (r=0,83), 3. Nellcor Pulsoxymeter N180: Met-Hb=(101-psO ₂ )· 0,9 (r=0,92).

Obwohl dieser Meßfehler für jeden Gerätetyp gut reproduzierbar ist, halten wir es dennoch nicht für zulässig, aus dem Abfall der psO ₂ auf das Ausmaß der Methämoglobinämie zu schließen. Die Bestimmung von Dyshämoglobinen muß daher immer in vitro, d.h. mit einem Co-Oxymeter erfolgen.

Abstract

During the last 15 years pulse oximetry has become a widely accepted method of monitoring during general and local anaesthesia. Pulse oximeters measuring with two wavelengths are considerably affected by dyshaemoglobin. At concentrations up to 30%, CO-Hb cannot be distinguished from O₂-Hb. Met-Hb, even in low concentrations, leads to a constant error of measurement; some authors recommended exploiting this for estimation of the Met-Hb concentration. To prove the aim of the present study was to test whether this error in measurement can be defined with one formula for different pulse oximeters.

Patients and methods. In a prospective, randomized, double-blind study, 171 non-smoking patients with healthy lungs (ASA 1–3) who had received a plexus block for hand surgery were investigated. After premedication with 3.75–15 mg medazolam p.o. each patient received a total of 6 lO₂ via a Hudson mask during the investigation. After 10 min the following pulse oximeters were put on the index finger: (1) Ohmeda BIOX 3700e, (2) Critikon Oxyshuttle, (3) Nellcor N 180. Simultaneously a venous blood sample was taken and analysed immediately with a Radiometer OSM3. The procedure was repeated 15, 30, 60 and 120 min after the plexus block. In 41 patients the plexus block was carried out with lidocaine (6 mg/kg body weight) and in 130 patients, with prilocaine (7 mg/kg body weight).

Results. There were no significant differences in age, sex and risk groups between the lidocaine and the prilocaine group. In the lidocaine group we were able to show that hyperoxic conditions can be maintained for 2 h with the method described. In the lidocaine group none of the pulse oximeters showed a psO₂ less than 99%. Our results show significant differences between the three pulse oximeters. Therefore, in contrast to the convention followed in the literatur, the relation between Met-Hb and psO₂ under hyperoxic conditions must be described with different formulas for each pulse oximeter as follows: (1) Ohmeda BIOX 3700e: Met-Hb=(101-psO₂)·0.6 (r=0.94); (2) Critikon Oxyshuttle: Met-Hb=(101-psO₂)·0.7 (r=0.83); (3) Nellcor N 180: Met-Hb=(101-psO₂) ·0.9 (r=0.92).

Discussion. Our results show that it is not possible to describe the connection between Met-Hb and psO₂ for all pulse oximeters with only one formula, but it is possible to set up different formulas with good correlations for each of the three pulse oximeters. The reasons for the different sensitivity are probably the different algorithms used by the manufacturers. In spite of the good correlations we can not recommend Met-Hb estimation by pulse oximetry measurement with two wavelengths, because the distinction of hypoxia and Met-Hb its not possible when hyperoxic conditions are not stable as they were in our controlled study. A low psO₂ measured in patients with normal arterial blood gases can be an indication of Met-Hb, but the exact measurement of dyshaemoglobin is only possibly by using a co-oximeter.