COMPARATIVE STUDY OF CONTINUOUS VERSUS INTERMITTENT SUBGLOTTIC SUCTION DRAINAGE TO PREVENT VENTILATOR ASSOCIATED PNEUMONIA IN INTUBATED PATIENTS

Compara ve Study Of Con nuous Versus Intermi ent Subglo c Suc on Drainage To Prevent Ven lator Associated Pneumonia In Intubated Pa ents. Akri Bajracharya, Lalit Kumar Rajbanshi. BJHS 2021;6(3)16. 16211625. Introduc on Subglo c secre on drainage reduces the incidence of ven lator associated pneumonia (VAP). The efficacy of Con nuous subglo c secre on drainage (CSSD) over Intermi ent subglo c secre on drainage (ISSD) for preven on of VAP is unknown.


INTRODUCTION
Ven lator associated pneumonia is one of the common consequences of mechanical ven la on in cri cally ill pa ents in intensive care unit. It is a nosocomial infec on which occurs a er 48 hours of endotracheal intuba on. It can occur as early as first 96 hours of mechanical ven la on. Early onset VAP is usually caused by community acquired bacteria like Hemophilus and Streptococcus. VAP which occurs more than 96 hours a er mechanical ven la on is 1 more common due to mul drug resistant microorganisms. The overall rate of VAP is 13.6 per 1000 ven lator days according to Interna onal Nosocomial Infec on Control 2 Consor um. The incidence of VAP ranges between 9 -27% in ICU despite wide use of preven ve measures and the 3,[4][5][6] crude mortality rate ranges from 25%-50%.
VAP also 2,7,8 increases health care costs by increasing stay in ICU. The presence of endotracheal tube or tracheostomy tube interferes with normal anatomy and physiology of 3 respiratory tract. Impairment of voluntary clearance of secre ons in intubated pa ents causes macroaspira on and microaspira on. Aspira on of oropharyngeal pathogens and leakage of subglo c secre ons containing bacteria around the endotracheal tube cuffs lead to entry of 2 bacteria into lower respiratory. Microbial invasion of the normally sterile lower respiratory tract and lung parenchyma causes VAP. Studies show that secre ons accumulate in subglo c area above cuff of endotracheal 9 tube before reaching lower respiratory tract. An endotracheal tube with specially designed dorsal lumen that opens immediately above the cuff is used for subglo c secre on drainage (SSD). Previous meta analyses have [10][11][12] shown that SSD decreased the rate of occurrence of VAP. Therefore, the use of SSD has been included in one of the components of VAP preven on bundles. In contrary, three other single center trials did not show decrease in the rate [13][14][15] of incidence of VAP. SSD can be done con nuously or intermi ently. Both CSSD and ISSD are used to prevent 10 VAP. In comparison to ISSD, CSSD is more efficient in decreasing the aspira on of subglo c secre ons into the lungs. However, there is paucity of the studies comparing the efficacy of con nuous subglo c suc oning over intermi ent subglo c suc oning in terms of decreasing VAP. The main aim of our study was to compare the efficacy of CSSD and ISSD in preven on of VAP. The secondary objec ves of our study were to determine the outcome of pa ents in terms of length of mechanical ven la on, number of days of ICU stay and mortality rate.

METHODOLOGY
This is a prospec ve compara ve study which was conducted in intensive care unit of Birat Medical College and Teaching Hospital from April to August, 2021. This study was approved by Ins tu onal Review Commi ee of Birat Medical College and Teaching Hospital (IRC-PA-115/2077th 78 and date of approval was 25 April, 2021) and wri en informed consent was taken from pa ent party. Eighty (80) pa ents who were intubated either in Emergency, ICU or Opera on Theatre and were in mechanical ven la on for more than 48 hours were included in this study. All the pa ents were intubated with Halyard Microcuff Subglo c Suc oning Endotracheal Tube. The pa ents with history of smoking, cough or sputum produc on and fever at the me of ICU admission and pa ents with chronic obstruc ve pulmonary diseases, asthma, respiratory failure and mul organ failure were excluded from the study. The included pa ents were randomly divided into two groups: Con nuous Subglo c Suc on Drainage (CSSD) and Intermi ent Subglo c Suc on Drainage (ISSD). In the con nuous group, subglo c suc on was delivered using a standard wall suc on unit at a nega ve pressure of 20 mm of Hg. Whereas the secre ons of pa ents in intermi ent group were drained intermi ently at an interval of 2 hours with the use of standard wall suc on unit at a nega ve pressure of 100 mm of Hg. Standard care of ICU was given in both groups. Every day spontaneous breathing trial was also given in every pa ent. The primary outcome of this study was to compare the incidence of VAP in two suc oning technique. At 48 hours of intuba on, pa ents were assessed for any new onset fever, any new changes in chest X-ray and purulent sputum produc on. If any one of them was present, gram stain and culture sensi vity of endotracheal tube secre ons were sent. Likewise, other parameters which include Glasgow coma score (GCS), PaO /FiO ra o, total leukocyte count, 2 2 mean arterial pressure (MAP), serum bilirubin, serum platelet count and serum crea nine were also noted at the me of admission, at 48 hours and 96 hours of endotracheal intuba on. All the included pa ents were assessed for development of VAP at 48 and 96 hours of endotracheal intuba on. In the mean me, we also consider secondary outcomes like length of mechanical ven la on, length of ICU stay, dura on of spontaneous breathing trials, re-intuba on, mortality rate and presence of mul organ failure. Data was collected and entered in Microso Office Excel. Then data was analyzed using Sta s cal So ware IBM SPSS sta s cs (version 23). The sample size of 80 was calculated on the basis of efficacy rates of con nuous and intermi ent 10 SSD according to the study done by Wang F et al where the es mated efficacy rates were assumed to be 20% and 55% respec vely. With an α error of 5% and power of study 80, total sample size was detected to be 80 (40 in each group). Con nuous data was presented as mean and standard devia on whereas categorical data was presented as frequency and percentage. Paired t test was used to compare mean for con nuous data and Chi square test was used for categorical data. P value < 0.05 was considered sta s cally significant.

RESULTS
A total of 100 pa ents were eligible ini ally in this study out of which 20 pa ents were excluded during the study period due to the various men oned exclusion criteria. Table 1 shows demographic features of pa ents in both groups. The mean age, sex and history of smoking were comparable between the two groups. Sta s cally significant number of the pa ents in CSSD group were intubated in ICU as compared to ISSD group pa ents (P=0.001) while a significant number of the pa ents were intubated in emergency in ISSD group. Likewise, the majority of pa ents are intubated elec vely (52.5%) in CSSD. While most of the pa ents are intubated on emergency basis (57.5%) in ISSD group. Table 4 shows outcomes of the pa ents in two groups. The length of mechanical ven la on was less in con nuous group (4.78 ± 2.50 days) as compared to intermi ent group (7.18 ± 2.09 days) with p value of 0.023 which is sta s cally significant. Similarly, the length of ICU stay was less in con nuous group (5.49 ± 3.16 days) than in intermi ent group (8.46 ± 2.06 days) with sta s cally significant difference (p value 0.014). Pa ents in both the groups developed comparable mul organ failure and spontaneous breathing trial was successful in comparable number of the pa ents. The pa ents in ISSD had significantly increased number of incidence of reintuba on (P=0.024). However, CSSD offered no mortality benefit over ISSD.
The table 2 shows clinical characteris cs of pa ents at the me of enrollment in this study. The ini al clinical features in terms of fever, X ray changes, mean arterial blood pressure, total leukocyte count, bilirubin, platelet PaO / 2 FiO were sta s cally comparable between the two groups 2 pa ents. SOFA score of 4 ± 2.32 and 4.10 ± 2.14 in CSSD and ISSD group pa ents respec vely showed that the severity of pa ents' condi on was similar in both groups. Table 3 shows the incidence of VAP. Total number of pa ents who developed VAP was 33. In the first 48 hours, 6 pa ents (15%) developed VAP in con nuous group and 7 pa ents (17.5%) had VAP in intermi ent group. While 9 pa ents (22.5%) had VAP in con nuous group and 11 pa ents (27.5%) developed VAP in intermi ent group in next 96 hours. Although the incidence of VAP was more in intermi ent group when compared with con nuous group, the results were not sta s cally significant. Nevertheless, sputum produc on was significantly increased in ISSD as compared to CSSD in 48 and 96 hours (p values 0.024 and 0.039 respec vely).    Table   Table 4: Outcome Variables

DISCUSSION
Endotracheal intuba on and prolonged mechanical ven la on are the leading cause of nosocomial pneumonia and ven lator associated pneumonia which has a major impact on the morbidity and mortality of the cri cally ill pa ents. Studies showed that secre ons accumulated above endotracheal cuff play an important role in 3,4,16 developing VAP.
Hence subglo c secre on drainage is used to remove the accumulated secre ons and assumed that it might decrease the incidence of VAP. The advantage of using subglo c suc on in decreasing VAP has been [10][11][12] established by some of the studies. Both con nuous and intermi ent SSD have shown to reduce the development of 10 VAP. The present study showed a comparable demographic profile and clinical features at the me of admission which made the comparison of the primary and secondary objec ves of the study uniform and unbiased by confounding factors. The severity of the disease in the pa ents in both the groups was similar. The present study looked into the incidence of ven lator associated pneumonia as a primary outcome. The incidence of VAP was 15% in con nuous SSD and 17.5% in intermi ent SSD at 48 hours in the study. Likewise, the incidence of VAP was 22.5% in con nuous SSD and 27.5% in intermi ent SSD at 96 hours. Rela vely, the incidence of VAP was less in con nuous subglo c suc oning technique though the difference was not sta s cally significant (p value 0.728 at 48 hours and p value 0.068 at 96 hours). We assumed that the con nuous suc oning of the tracheal secre on led to effec ve clearance of the subglo c secre ons thus decreasing the probability of developing VAP. Similar findings were 17 observed in a study conducted by Fujimoto et al in which the incidence of VAP was 26.7% in con nuous SSD and 43.8% in intermi ent SSD. Although the study showed decreased incidence of VAP in con nuous group than in intermi ent group, it was not sta s cally significant (p value 0.320) and the author assumed that the reason behind the cause was due to early extuba on leading to exclusion of the par cipants from the study. The study by Fujimoto et al had higher VAP incidence in both the group as compared to our study. The reason behind it was assumed to be broad diagnos c criteria. Over the years, the diagnos c criteria for VAP has been more precise. Hence, the incidence of VAP is reduced in both groups at 48 hours and 96 hours in our study. An overview of systemic reviews and an updated metaanalysis showed that subglo c secre on drainage (SSD) 18 significantly reduced the incidence of VAP. Similarly, a 19 review ar cle showed that SSD primarily decreases the incidence of early-onset VAP. It also showed that SSD had no effect on late-onset VAP. It stated that none of the published studies were experimentally designed to demonstrate, in inten on to treat, any effect on late-onset VAP. Among 20 20 RCTs , four high quality studies were associated with reduc on in incidence of VAP. Overall, it showed rela ve risk of 0.55 for the incidence of VAP in SSD versus non SSD. 21 Likewise in a meta-analysis , SSD reduced VAP rate by nearly half as compared to ET without SSD. Another meta- 22 analysis demonstrated that SSD reduced incidence of VAP by 50%. 23 In a mul center trial , the intermi ent SSD significantly reduced the incidence of early onset and late onset VAP. It also showed that 11 cases of VAP could be avoided in every 100 pa ents treated with SSD. In previous studies, con nuous SSD also showed decreased incidence of 13,15,24 25 VAP.
In contrary to our study, a metaanalysis showed no advantage of CSSD over ISSD in the incidence of VAP. The differences in methodology, end points and the included study groups were assumed to be the poten al causes for not being able to find the difference between the two different types of SSD. In our study, the pa ents with con nuous drainage had shorter dura on of mechanical ven la on and ICU stay as well, as compared to pa ents with intermi ent drainage technique. Though sta s cally not significant, the incidence of VAP was rela vely less in pa ents with con nuous drainage and it was obvious that con nuous drainage cleared the subglo c secre ons more effec vely which might have led to less number of days in mechanical ven la on and decreased length of ICU stay. This finding in 17 our study has been supported by Fujimoto Hiroka et al. They demonstrated a significant decrease in dura on of mechanical ven la on and length of ICU stay with the use of 17 con nuous drainage system. In contrary to our study, there was no advantage of CSSD over ISSD in terms of dura on of mechanical ven la on, ICU stay and mortality rate in a 25 meta-analysis. The con nuous drainage of the subglo c secre on had led to significant decrease in the incidence of reintuba on. The incidence of reintuba on in CSSD is less as compared to ISSD in our study. In our study, there is no mortality benefit of con nuous drainage over intermi ent drainage technique. The mortality of cri cally ill pa ents is affected by numerous clinical and pathophysiological condi ons like severity of the diseases, mul organ failure, organ support system. VAP is one of the a ributable clinical condi on for the mortality of cri cally ill pa ents. As there is no benefit decreasing the incidence of VAP in the present study, the mortality benefit remained comparable between the two groups. Similar to our study, Fujimoto Hiroka et al also failed to show mortality benefit of con nuous drainage over intermi ent drainage 17 system. Mao Z et al showed that there is an associa on of SSD and decrease in dura on of mechanical ven la on. But it failed to show a correla on between SSD and decrease in 20 ICU stay, mortality and reintuba on. Whereas, a significant difference was seen in mortality when compared with SSD 18 versus non SSD interven on in a study. But it did not show any associa on between SSD technique and decrease in length of ICU stay and dura on of mechanical ven la on. A mul center trial also showed no difference in dura on of mechanical ven la on, length of ICU stay and mortality rate 23 between con nuous and intermi ent drainage.

CONCLUSION
The con nuous drainage of subglo c secre ons has no added benefit over intermi ent drainage system in terms of decreasing the incidence of ven lator induced pneumonia and mortality. However, the use of con nuous subglo c suc on in clearing the secre on significantly reduced the dura on of mechanical ven la on and length of ICU stay.

LIMITATIONS
This study has several limita ons. First, the sample size was less in our study. Hence, significant difference in incidence of VAP could not be found between CSSD and ISSD. We found significant differences in terms of secondary outcomes only. Second, there is heterogeneity in terms of site of intuba on between two groups. Third, SSD is an obvious clinical manipula on. Hence, it could not be blinded between doctors and nurses leading to unavoidable performance bias.

RECOMMENDATION
SSD should be used in ICU inorder to decrease the incidence of VAP. CSSD is be er than ISSD inorder to reduce length of mechanical ven la on and dura on of ICU stay.