Suspected Malfunction of the SAF-T-FILL™ Valve Assembly of the Suprane® (Desflurane, USP) Refill Bottle

Desflurane is an inhaled anesthetic with unique physical properties. At 760 mm Hg atmospheric pressure, desflurane boils at 23.5°C (1). Consequently, a new supply and delivery system was designed for the introduction of this drug into its drug-specific vaporizer–the Tec 6 desflurane vaporizer (Ohmeda, Madison, WI) (2). Desflurane is supplied in brown, plastic-coated bottles that contain an integrated SAF-T-FILL™ system (Zeneca Pharma, Inc., Mississauga, Ontario). Each bottle contains 240 mL of desflurane (3). During the course of examining 156 desflurane refill bottles from 26 shipping boxes, we discovered two incompletely filled bottles. We assumed that the bottles had been partially filled at the time of manufacture. Nonetheless, we examined these two unusual desflurane bottles and compared them with the remaining 154 bottles. Methods We examined the clear plastic security seals of 156 desflurane refill bottles to confirm their integrity, and then removed them; the bottles were weighed at 20°C to document their full weights. The white screw-on bottle caps and bayonet fitments (bottle probes) were scrutinized. The refill bottles were then used to fill the Tec 6 desflurane vaporizers as required. Finally, the empty bottles were vented to ensure that any residual desflurane had been removed. After the white bottle caps were replaced, the bottles were weighed at 20°C to document their empty weight. To calculate the volume of the desflurane the refill bottles contained, the difference between the full weight and the empty weight of the bottles was divided by the specific gravity of desflurane at 20°C (1.465 g/mL) (4). To determine if the weights of the 154 full refill bottles conformed to a Gaussian distribution, the Kolmogorov-Smirnov (KS) normality test was used. This test quantifies the discrepancy between the distribution of the data collected and an ideal Gaussian distribution. To compute a P value, the Dallal and Wilkinson approximation to Lilliefors’ method was used. Large P values (P > 0.10) indicate that the data passed the normality test and, therefore, were Gaussian. For the remaining two study bottles, the Grubb’s method for assessing outliers was applied to the extreme outlier (5,6). A P value of <0.01 confirmed that the outlying bottle differed significantly from the sample group. Results All of the clear plastic security seals were found to be intact. None had been removed or damaged before the weighing of the refill bottles. The white screw-on caps and bayonet fitments of the bottle probes had no visible damage. The mean weight of the 154 full bottles was 617.5 g (SD ± 4.2 g). These refill bottles conformed to a Gaussian distribution with a KS distance of 0.072 and a P value exceeding 0.10. The two unused partial bottles had full weights of 528.6 and 559.8 g (mean weight 544.2 g). These outliers were significantly beyond the normal distribution of the study sample (Figure 1). The Grubb’s test was significant at P < 0.001, confirming that the extreme outliers differed from the remaining 154 refill bottles.Figure 1: The weights of full bottles of Suprane® (Desflurane, USP).The mean empty weight of the 154 bottles was 259.1 g (SD ± 4.1 g). The empty weights of these bottles conformed to a Gaussian distribution with a KS distance of 0.054 and a P value exceeding 0.10. The empty weights of the two outlying partial bottles were 263.6 and 253.0 g (mean weight 258.3 g), confirming that the empty weights of the partial bottles belonged to the same Gaussian distribution as the other 154 empty bottles. This confirmed that the two outlying bottles differed from the remaining 154 refill bottles because of the reduced amount of desflurane that they contained. The mean volume of desflurane that was contained in the 154 full bottles was 244.6 mL (SD ± 0.9 mL). The volume that was contained in the two outlying refill bottles was markedly less at 180.9 and 209.4 mL. During the course of the study, while examining one of the unused partially full refill bottles, an emergency distracted me (DTJ), and I was compelled to place the bottle on its side in an empty cardboard box, forgetting to replace the white screw-on cap. On my return 30 min later, I discovered that the bottle probe was encrusted with frost (Figures 2 and 3). The frost was removed and the screw cap replaced. The next day, the screw-on cap was removed, and the bottle was placed on its side again with the same results.Figure 2: Suprane® (Desflurane, USP) bottle probe encrusted with frost, superior view.Figure 3: Suprane® (Desflurane, USP) bottle probe encrusted with frost, lateral view.Discussion The integrated SAF-T-FILL™ system offers several advantages. Not only does it prevent the inadvertent introduction of desflurane into a traditional vaporizer, the spring valve assembly helps to minimize the wastage of volatile agents during filling of the vaporizer. The spring valve assembly acts as a back-up system to prevent the evaporative loss of desflurane during storage of the refill bottle when the operator fails to replace the screw-on cap. Nonetheless, wastage of volatile agents associated with the use of keyed filling systems has been described (7,8). Uncles et al. (9) reported in 1994 their suspicions that the spring valve assembly of the SAF-T-FILL™ system could fail and permit the escape of desflurane. Our serendipitous observation confirms that the SAF-T-FILL™ spring valve assembly of the bayonet fitment rarely malfunctions. This permitted desflurane to escape from the refill bottle and evaporate and cool the bottle probe sufficiently to cause the crystallization of water vapor on the bottle probe. This stresses the importance of tightly securing the white screw-on cap after the refill bottle has been used to fill a Tec 6 desflurane vaporizer. Failure to replace the screw-on cap could result in exposing operating room personnel to an avoidable pollution hazard should the SAF-T-FILL™ spring valve assembly malfunction. In an enclosed and poorly ventilated space, the exposure could be significant. Furthermore, this preventable loss of desflurane, an expensive drug, deprives the purchaser of value. The white screw-on cap must be securely replaced to prevent the loss of desflurane to the atmosphere, should the spring valve assembly malfunction. This is particularly important when only a portion of the contents of a refill bottle has been dispensed. Our observation that the average refill bottle dispenses 244.6 mL (SD ± 0.9 mL) confirms the report by Uncles et al. (9) that the suppliers of desflurane routinely increase the fill volumes of the refill bottles to exceed 240 mL. This was done in an effort to cover the inevitable and irrecoverable loss of drug resulting from the residual volume remaining in the refill bottle after it has been completely emptied into the sump of the Tec 6 vaporizer. Thus, the manufacturer has ensured that the consumer receives value for their purchase. However, if the SAF-T-FILL™ spring valve assembly fails, the potential loss of drug could still deprive the consumer of value. The discovery of new, never-before-used desflurane refill bottles that were partially filled attracted our interest. Because the security seals and screw-on caps appeared intact and undamaged, our initial reaction was that the bottles had been partially filled at the time of manufacture. The discovery of a malfunctioning spring valve assembly of the bayonet fitment of the bottle probe suggested an alternative explanation. Although no visible defects in the screw-on cap or the bottle probe were detected, it is conceivable that a malfunctioning spring valve assembly permitted the loss of desflurane vapor through either the bottle cap or the bayonet fitment. It has been our experience that the manufacturer will graciously exchange desflurane refill bottles that are suspected of being defective (10,11).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.