Analgesia & Resuscitation : Current ResearchISSN: 2324-903X

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, Analg Resusc Curr Res Vol: 3 Issue: 4

Incidence of PONV in Maxillofacial Surgery: Comparison of High-Flow and Low-Flow Anesthesia

Agreta Gecaj-Gashi1*, Shaqir Uka1, Mergime Prekazi-Loxha2, Zorica Nikolova3, Paul Zilberman4 and Antigona Hasani1
1Clinic of Anesthesiology & Intensive Care, University Clinical Centre of Kosova, Prishtina, Republic of Kosova
2Clinic of Maxillofacial Surgery, University Clinical Centre of Kosova, Prishtina, Republic of Kosova
3Clinic of Anesthesiology & Reanimatology and Intensive Care, University Clinical Centre of Skopje, Republic of Macedonia
4Anesthesia Department, Har Hatzofim Campus, Hadassa University Medical Center, Israel
Corresponding author : Agreta Gecaj- Gashi
Clinic of Anesthesiology and Intensive Care, University Clinical Center of Kosova, 10000 Prishtina, Republic of Kosova
Tel: 0037744137404
E-mail: [email protected]
Received: August 21, 2014 Accepted: November 19, 2014 Published: November 25, 2014
Citation: Gecaj-Gashi A, Uka S, Prekazi-Loxha M, Nikolova Z, Zilberman P, et al.(2014) Incidence of PONV in Maxillofacial Surgery: Comparison of High-Flow and Low-Flow Anesthesia. Analg Resusc: Curr Res 3:4. doi:10.4172/2324-903X.1000128


Incidence of PONV in Maxillofacial Surgery: Comparison of High-Flow and Low-Flow Anesthesia

Postoperative Nausea and Vomiting (PONV) still remains the “big little challenge” for anesthesiologists in their everyday practice. The aim of this study was to compare the incidence and severity of PONV, antiemetic requirement and patient satisfaction, after high-flow versus low-flow sevoflurane anesthesia without antiemetic prophylaxis.

Keywords: PONV; Maxillofacial surgery; Anesthesia


PONV; Maxillofacial surgery; Anesthesia


Postoperative Nausea and Vomiting (PONV) is one of the most frequent unpleasant side effects of general anesthesia, and still remains a “big little problem” [1] for patients, but at the same time, a “big little “challenge for anesthesiologists.
The overall incidence of PONV ranges from 20% to 30% [2]. In “high-risk” patients this incidence remains very high at approximately 70% [1]. Factors influencing PONV development according to Apfel et al. [3] are: female gender, age, nonsmoking status, previous history of PONV or motion sickness, and use of intra- or postoperative opioids.
With the recent revolution in anesthesia technical developments, the low-flow anesthesia is becoming slowly popular among anesthesiologists. The advantages of low-flow anesthesia are: more economical, as there is a significant reduction of costs; climatization of the inspired gases; and reduced pollution with inhalational anesthetic [4-7].
During low flow techniques of anesthesia, because of aceton accumulation, the patients have a significantly greater concentration of acetone in their blood, and this can cause a increased incidence of PONV [8].
The aim of this study was to compare the incidence and severity of PONV, antiemetic requirement, and patient satisfaction, after high-flow versus low-flow sevoflurane anesthesia without antiemetic prophylaxis.


After obtaining approval from our hospital ethical committee and written informed consent from all participants, 44 patients, ASA physical status I–II, age 18-65 years, weighing between 40 and 90 kg, who were scheduled for elective maxillofacial surgery under general anesthesia that was expected to last not less than 1 hour and not more than 2 hours, were enrolled in this pilot, prospective, randomized, double-blinded study. The patients were randomly allocated via a computer-generated random number list into two groups: group HFA (n=32) using high-flow anesthesia and group LFA (n=32) using lowflow anesthesia with sevoflurane.
Exclusion criteria were Apfel score>II, antiemetic use within 24 h before surgery, chemotherapy use within 4 weeks or radiotherapy within 8 weeks, smoking, allergy to any of the study drugs, motion sickness, epilepsy, mental retardation, and women who were menstruating, pregnant, or lactating. All patients received oral diazepam (Diazepam, Actavis UK Ltd) 10 mg in the evening before operation.
Our primary end-point was incidence and severity of PONV in early period 0-6 hours, while secondary end-point was incidence and severity of PONV in late period 6-24 hours.
On arrival in the operating room (OR), intravenous access was obtained with an 18-gauge IV cannula, and then standard monitoring applied: Datex -Ohmeda S/5 (TM) (Helsinki, Finland), electrocardiogram, noninvasive blood pressure, pulse oxymeter. The baseline parameters were noted. All patients received midazolam (Dormicum®, F. Hoffman-La Roche Ltd Basel, Switzerland) 0.03 mg/ kg iv 5 minutes before induction. All were preoxygenated with 100% O2 for 3 minutes. Anesthesia was induced with propofol 2.5-3.0 mg/ kg, fentanyl 1.5-2.5 μg/kg, and succinylcholine 1 mg/kg to facilitate intubation. After intubation, the fresh gas flow was set to 4 L/min (O2 2 L/min, N2O 2 L/min) and sevoflurane 1-1.5 MAC (2.0-3.0%) for 10 min in LFA group, whereas the HFA group continued with the same flow during all periods of anesthesia.
Mechanical ventilation was performed in both groups by using a rebreathing circuit (Dräger® Fabius GS, Lübeck, Germany). The ventilation was adjusted to keep SaO2 > 95% and to maintain ETCO2 at 35 to 45 mmHg.
After the high-flow phase, when the target gas concentrations 0.8-1.0 MAC (end-tidal) has been achieved, FGF in the HFA group remained unchanged at 4 L/min, whereas this was reduced to 1.0 L/ min in the LFA group, with the relative proportion of oxygen and nitrous oxide titrated to maintain the inspired oxygen fraction (FiO2) above 0.35.
Anesthesia was maintained with sevoflurane (Sevorane®, Abbott Lab, North Chicago, ABD; end-tidal concentration 0.8-1.0 minimum alveolar concentration [MAC]) and supplemental bolus doses of fentanyl IV (1 μg/kg) and atracurium for muscle relaxation.
Insertion of a nasogastric tube and gastric suction were not performed. Throat packing was used in all patients to prevent the possibility of blood and secretion to pass on to the stomach.
In either group, the sevoflurane vaporizer was closed about 10 minutes before the end of surgery, whereas nitrous oxide supply was ceased 5 to 10 minutes before the definitive extubation and the oxygen flow increased to about 5 L/min to wash out the anaesthetic. At the end of the procedure, residual neuromuscular block was reversed with atropine and neostigmine administration, and extubation was performed awake. After that, the patients were transported to the postanaesthetic care unit (PACU). Vital signs (blood pressure, heart rate, SpO2) were recorded at 5-min intervals throughout surgery and at 10-min intervals in PACU until the patient was fully alert.
In both groups, ketorolac (Eumat, Epifarma, Italy) 30 mg IV was administered 15 min before the end of surgery and after that in the PACU as needed for postoperative pain.
The incidence and severity of PONV was evaluated for 24 hours postoperatively based on this scoring system: 0=no emetic symptoms, 1=nausea, 2=vomiting. Nausea severity was recorded on a 4-point categorical Likert scale: 0=none, 1= mild, 2=moderate, 3=severe. A complete response (CR) was defined as no PONV and no need for rescue antiemetic.
Rescue antiemetic-metoclopramide 10-20 mg IV was administered when the PONV score was greater than 1 or when Likert scale was 2-3 lasting >15 min.
In PACU, a nurse who was blinded to the intraoperative management recorded the number of nausea and emetic episodes, the time each one occurred, and the requirement of rescue antiemetic medication. PONV was recorded in two stages: early postoperative period (0-6 h) and late postoperative period (6-24 h).
Patients were discharged from PACU to the surgical ward when they were fully awake and oriented, had stable vital signs and minimal pain (<3 on a 0–10 VAS scale), and were not experiencing any side effects. PONV assessments were made and recorded in the surgical ward by the nurse on duty who was also blinded to the method used.
Patients rated their satisfaction with the control of PONV using a 5-point scale (1 = very satisfied; 2 =somewhat satisfied; 3 = neither satisfied nor dissatisfied; 4 = somewhat dissatisfied; 5 = very dissatisfied) approximately 24 h after anesthesia.
For testing of all categorical data, we used Fisher exact test, χ2 test, and Kruskal Wallis, whereas for parametric data, t test was applied. Spearman’s correlation was obtained to evaluate the relationship between Apfel score and PONV. P<0.05 is considered significant.


There were no significant differences between the groups with respect to demographic data, ASA score, and Apfel score (Table 1).
Table 1: Demographic and baseline characteristics of patients.
Efficacy data are summarized in Table 2. There was no significant difference among the groups in the incidence of moderate to severe nausea (2-3 Likert scale) in the HF group (40.9%) compared to the LF group (27.3%) in the early postoperative period (0-6 h) (P=0.525). During late postoperative period (6-24 h), no significant difference was found (P=0.698) between group HF (22.7%) and group LF (13.6%).
Table 2: Incidence of PONV, rescue antiemetic, and satisfaction between groups.
Incidence of vomiting or retching (score 2) in the early postoperative period (0-6 h) in group HF was 31.8% compared with 22.7% in group LF (P=0.736). In the late postoperative period (6-24 h) 18.2% in group HF suffered from vomiting or retching, whereas 9.1% suffered from vomiting or retching in group LF (P=0.664), which was not statistically significant (Figure 1,2 and 3).
Figure 1: Moderate to severe nausea by groups.
Figure 2: Vomiting or retching Score by groups.
Figure 3: Rescue antiemetic administration by groups.
The analysis shows no significant difference between groups regarding rescue antiemetic administration (P=0.736) in the early and late postoperative periods (P=0.412). The number of patients who were very satisfied with PONV management approximately 24 h after anesthesia was superior in the LF group compared with the HF group, but with no significant difference between groups (P=0.685).
A significant correlation between Apfel score ≤ 2 and incidence of nausea and vomiting was obtained in the early postoperative period in both groups, whereas there was less correlation in both groups in the late postoperative period (Table 3).
Table 3: Correlation between Apfel score = 2 and incidence of PONV.


The incidence of vomiting or retching and moderate to severe nausea, in early and late postoperative periods, was higher in the HF group than the LF group, but with no significant difference.
In patients undergoing maxillofacial surgery, swallowed blood and secretions stimulate the gag reflex and may increase the incidence of PONV, so we used a throat pack in all patients in order to reduce the possibility of blood and secretion to pass on stomach.
There are some reports with various incidences of PONV after orthognatic surgery, from 7% [9] to 40% [10] with antiemetic prophylaxis and up to 83% without antiemetic prophylaxis [11]. In our previous study for PONV in orthognatic surgery, we found an incidence of 9.0% with dexamethasone and 27.2% with metoclopramide prophylaxis in the early phase (0-6 h) [12].
The results of our study correspond with the data mentioned above, wherein the incidence of nausea was 40.9% in the HF group and 27.3% in the LF group, and incidence of vomiting was 31.8% in HF group compared with 22.7% in the LF group in the early phase without antiemetic prophylaxis.
There are many studies that confirm that the cause of PONV is multifactorial. These factors may be related to the patient, the surgical procedure, or the choice of anesthetics [13,14].
A nonsmoking status is considered as a strong PONV risk factor [15-17]; all patients in our study in either group were nonsmokers, and we consider that this was one important contributing factor in our results. There is evidence that fentanyl, when used in analgesic doses during anesthesia, does not increase the incidence of PONV [18,19]. We used fentanyl in both groups in analgesic doses accordingly.
Sinclair et al. [15] found that the prevalence of PONV increased from 2.8% in patients whose surgery lasted less than 30 minutes, to 27.7% in patients with a surgical duration of 150-180 min. The anesthesia time in our study was from 60 to 120 minutes, so this was also a contributing factor in our incidence. According to one study, there is no relationship between the low oxygen concentration and PONV [20], although in our study, FiO2 has never fallen below 35% in either group.
Regarding anesthesia as a contributing factor in PONV, nitrous oxide was considered a potential emetogenic agent that contributes significantly to PONV [21,22]. There are many studies that confirm the positive relationship between N2O and PONV [23-25], and others who suggest that N2O has no impact on the incidence of PONV [26,27].
It has been suggested that a proemetogenic effect of volatile anesthetics must be considered the main trigger of PONV in the early phase (postoperative period of 0-2 h) [28], which is consistent with the results from our previous study [29].
Dashfield et al. reported an incidence of nausea of 10% using high concentrations of sevoflurane, but they used a nonrebreathing system [29], whereas Smith and Thwaites used low-flow anesthesia with a rebreathing system and CO2 absorption, and reported a 30% incidence of PONV [30]. In another study, a high incidence of nausea (78%) was reported following 8% sevoflurane induction and maintenance at low fresh gas flows (2 L/min) [31]. Fredman et al. also reported a high incidence of nausea (56%) following a lower dose of sevoflurane for induction and maintenance of anesthesia using low fresh gas flows [32].
Because sevoflurane can be degraded to formaldehyde when passed through partially exhausted soda lime, it is assumed that inhalation of formaldehyde plays a contributing factor in PONV when a rebreathing system is used; rather than delivering a dose of sevoflurane [33], we used fresh soda lime for every patient.
The number of patients who were very satisfied with PONV management approximately 24 h after anesthesia was superior in the LF group compared with the HF group, but with no significant difference between groups.
We may assume that the difference of PONV between groups, in favor of the LF group, can be a result of the lowest concentration of N2O and sevoflurane delivered to the patient, humidification, and warming of the inspired gases, otherwise the highest satisfaction scale among the patients undergoing low-flow anesthesia cannot be explained.


Low-flow sevoflurane anesthesia has a tendency to be more advantageous regarding the incidence of PONV and satisfaction rate, but with no statistically significant differences in groups comparisons.


This is a pilot study with relatively small number of participants, so we intend to do further studies with focus in low flow anesthesia without N2O use.


Track Your Manuscript