The impact of implementation of a hysterectomy enhanced recovery pathway on anesthetic medication costs

The results of this study demonstrate that implementation of our hysterectomy ERP was not associated with an increase in perioperative anesthetic medication costs. Instead, on average the median total cost in the post-ERAS group was $40.76 USD lower than the median total cost in the pre-ERAS group. This corresponded to a 12.5% reduction in cost between the two groups. Patients in the post-ERAS group experienced less complications and attained their clinical milestones more frequently. Finally, the median hospital LOS was 2 days shorter in the post-ERAS group. This reduction in hospital LOS is similar to that described in other studies, validating our protocol’s effectiveness [7,21]. Implementation of our ERP was associated with excellent clinical outcomes and was no more costly than our pre-ERAS anesthetic technique. This indicates that perioperative providers can adhere to ERAS principles and deliver high quality care without incurring additional expense. These results are important for financial forecasting and outcomes analysis, especially as the US healthcare system transitions to a value-based care model. There are several potential reasons for our study’s findings.

Decrease in practice variation through standardization may explain the reduction in anesthetic medication cost following implementation of our protocol. More providers were likely to employ a multimodal analgesic regimen that included preoperative administration of inexpensive oral medications instead of administering expensive iv. agents intraoperatively. In the pre-ERAS group, only 9.5% of patients (4/42 patients) received oral acetaminophen while 76.2% of patients (32/42 patients) received iv. acetaminophen. Following ERP implementation, the percentage of patients who received oral acetaminophen increased to 95.2% (40/42 patients) while the percentage of patients who received iv. acetaminophen decreased to 23.8% (10/42 patients). This administration route standardization helps to explain our findings given that 1 g of oral acetaminophen cost $0.02 USD while 1 g of iv. acetaminophen cost $36.06 USD. This is shown in Table 4 in the significant reduction in the cost of nonopioid analgesics in the post-ERAS group. Table 4 also displays the effect of standardizing our antibiotic administration practice and the resulting costs savings the followed. Finally, the impact of standardization can be directly observed in Figure 1 and Figure 2; providers exhibited less variation in the amount of administered opioids and colloids as a result of implementation of our ERP.

In a study using regression models to evaluate sources of variation in anesthetic drug costs, Wanderer et al. examined anesthetic records for patients undergoing eight types of surgical procedures at two institutions [22]. Although the majority of variation could not be described by their models, the largest sources of variation included attending anesthesiologist, procedure type and case duration. While procedure type and case duration are not easily modifiable sources of variation, practice patterns among attending anesthesiologists are open to modification. In their study, with all other variables held constant, the average change in cost between attending anesthesiologists ranged from a cost decrease of $41.25 USD to a cost increase of $95.67 USD when compared with the provider with the median value for mean cost-per-case [22]. Depending on the scale of operations, the magnitude of cost savings from practice modification could be considerable. Distinct from providers, that unexplained factors represented the greatest amount of observed variation in their study suggests there exists considerable variation resulting from either unmeasured influences or random per-case variation. We observed a similar pattern in our data. The results of our regression analysis were unable to identify a single variable that accounted for the difference in cost between our groups. Overall, protocol-driven care may help to reduce hidden sources of variation (and subsequently cost), even if the mechanisms by which they do so are not immediately apparent.

There was an increase in the number of patients in the post-ERAS group who received sevoflurane as their volatile anesthetic agent while significantly more patients in the pre-ERAS group received desflurane. Within ERAS pathways for this patient population, there are no recommendations regarding the use of specific volatile agents. In order to facilitate a faster emergence, some guidelines suggest the use of short-acting inhalational agents such as sevoflurane or desflurane [23,24]. We did not advocate for the use of any one agent as these are the two volatiles that are routinely used at our institution. There are several methods to quantify volatile anesthetic consumption. One method involves weighing the vaporizer before and after use and taking the difference as the consumed amount. This practice requires advanced planning and a precise scale with a wide range of measurements [25]. Volatile anesthetic costs can be calculated using the vaporizer concentration, fresh gas flow rate, time, molecular weight, density and cost of the specific agent [25,26]. Unfortunately, our electronic medical records did not delineate these individual metrics. Given the historical nature of our study we were unable to calculate exact volatile agent usage. Consequently, we assigned a uniform volatile cost of $130.47 USD to patients who received desflurane and a $74.49 USD cost to patients who received sevoflurane. These values represented the costs for a full bottle of the respective agent. We classified each patient as having received only one volatile agent for their anesthetic. In cases where patients received multiple agents during the duration of the case, they were assigned the agent that they received for the longest amount of time. As shown in Table 3, even after removing all volatiles from our analysis, the median cost in the post-ERAS group was $46.06 USD lower than the pre-ERAS group, corresponding to a 19.5% reduction in cost. Therefore, we believe it was unlikely that the choice of volatile agent served as a major cost saving mechanism between the two groups.

There were procedure-related characteristics that may have contributed to our findings. Patients in the post-ERAS group underwent shorter anesthesia and surgery times compared with patients in the pre-ERAS group. With more time available, it stands to reason that providers could administer more medications and iv. fluids, thus increasing the total cost of the delivered anesthetic. However, when we analyzed our data on a per minute basis, our findings remained similar. The median cost per minute of anesthesia time was $0.06 USD lower in the post-ERAS group, corresponding to a 5.1% reduction in cost. Likewise, the median cost per minute of surgery time was $0.08 USD lower in the post-ERAS group, corresponding to a 5.0% reduction in cost.

It is unlikely that patient-related characteristics impacted our results. While there are sex-gender differences in the dose response to several anesthetic medications and the overall response to pain, all of the patients included in our study were female [27]. The normal aging process impacts the pharmacokinetics and pharmacodynamics of many anesthetic agents and in general, elderly patients have increased drug sensitivity. As such, providers are advised to administer smaller doses of medications when caring for this population [28]. This recommendation was unlikely to have influenced our results as patients in our study were of similar age. As with aging, the physiologic and anthropometric changes associated with obesity affect the pharmacokinetics and pharmacodynamics of anesthetic medications [29]. Drug doses are typically calculated using a patient’s actual bodyweight and patients in both groups were of similar bodyweight [30]. The ASA-PS classification system is commonly used by perioperative providers to estimate preoperative health status. The system includes six categories with increasing levels of patient impairment [31]. Although it is subjectively assigned, the ASA-PS has strong independent associations with postoperative complications and mortality across procedures [32]. We believe patients in our study were of similar preoperative health as there were no significant differences in ASA-PS classification scores for patients in the two groups.

Our study has several limitations. The study population was relatively small and consisted of a homogenous group of patients undergoing a specialized procedure at a single institution. Despite the recommendation for minimally invasive techniques in our ERAS protocol, a considerable number of procedures requiring an open surgical technique are still performed at our institution. This is most commonly done when the patient has a history of prior abdominal surgery or extensive disease, making a minimally invasive approach more technically challenging. This patient selection characteristic may limit the generalizability of our results. We were unable to accurately provide costs for volatile agent administration like some other pharmacoeconomic studies, however, our results remained similar even after removing volatiles from our calculation. We focused exclusively on medications that were administered in the preoperative and intraoperative phases and did not include any agents that were administered in the PACU. While we quantified the degree of patient health status using ASA-PS classification, it is possible that delineating a more detailed list of comorbidities may have better differentiated our two groups. Finally, when analyzing the impact of an initiative containing multiple interventions, it is difficult to discern which intervention yields positive results and which does not.

The results of this study demonstrate that implementation of our hysterectomy ERP was associated with a significant reduction in hospital LOS and a significant reduction in the total perioperative anesthetic medication cost. Our data indicate that ERAS protocols can improve clinical outcomes and lead to cost savings in these phases of care. The results provide evidence that ERAS programs can add unique value in the preoperative and intraoperative settings, separate from their impact on rates of postoperative complications and total cost of care.

AD Street, JM Elia and BT Romito did study design; AD Street, JM Elia, BT Romito, MM McBroom, AJ Hamilton, JE Grundt and JMN Blackwell worked on data analysis and interpretation of results; manuscript writing and editing.

The authors are grateful to the members of the Division of Gynecologic Oncology at the University of Texas Southwestern Medical Center for their teamwork in making our ERAS protocol successful.

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

The authors state that they have obtained appropriate IRB approval or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations. The requirement for written informed consent was waived by the IRB as this was a historical study.