Healthcare-associated pneumonia represents up to 27% of all ICU infections, and those types of pneumonia could also be a complication of severe lung contusion due to chest trauma. Of that pneumonia, 90% occur in mechanically ventilated patients, costing approximately 10,962.5 United States dollars/year [1, 2]. Severe sepsis could be a cause of ventilator-associated pneumonia (VAP) or may be one of its major systemic complications [1-3]. 65% of patients with sepsis are associated with pneumonia, especially VAP. Both conditions are frequently found in most of the ventilated ICU patients. Mortality rate for VAP with sepsis is 18%–30% [2-4]. Some evidence was raised by many researchers that diagnostic bronchoscopy may improve VAP related outcomes and thus improve or even stop the cascade of septic reaction, although this is also disputed in the literature [2-5]. But still source control through the removal of mucus plugs from the endotracheal tube and small airway could improve outcomes and put this maneuver as one of the respectable lines of management of VAP with sepsis [6, 7]. On the basis of previous reports, our hospital established the use of the aggressive bronchoscopic toilet for VAP patients with severe sepsis as a standard line of management. We have developed an informal, clinician-driven protocol involving bronchoscopic diagnosis with broncho-alveolar lavage (BAL), and serial therapeutic bronchoscopies for patients believed to be at high risk. The number of serial bronchoscopies is based on the appearance of the airway and the volume of purulent secretions in the previous evaluation. Patients, who were believed by the treatment team to be at high risk for a respiratory decline from VAP and showed an aggressive septic reaction, especially those who had thick secretions or difficulty clearing their airways, were generally treated in this manner. Based upon our experience with this practice, we conducted this study to monitor the effect of this maneuver in lowering mortality in those with VAP and severe sepsis, improve clinical resolution of VAP in septic patients, accelerate weaning from the ventilator, control the cascade of sepsis and shorten the duration of ICU stay.

This prospective cohort study was conducted on 200 patients admitted to our intensive care unit in King Abdul-Aziz Specialized Hospital, Taif, KSA, between July 2020 and December 2021 in the surgical ICU. King Abdul-Aziz Specialized Hospital research and Ethical committee approved the project and written consent from all patients was taken from the first-degree relatives of the patients.

Inclusion criteria: Adult patients aged >18-<65 years, those who were intubated and ventilated due to ARDS with a severe lung infection and/or traumatic lung contusion, those who had hypoxic index less than 200 [PO₂/FIO₂ less than 200], and had bilateral parenchymatous lung infiltration on their chest X ray, those who showed PH > 7.30 on their ABG, and those who had respiratory rate >25 min. All those patients were ventilated for 4 days with CMV with respiratory rate 12/min, PEEP 5 cm/ H₂O, FIO₂ adjusted to maintain their arterial oxygen saturation above 90%, and sedated with both fentanyl and midazolam intravenous infusion to adjust their sedation level to achieve Richmond Agitation-Sedation Scale (RASS) -2 to -3 as illustrated in Table 1 [8]. All patients received broad spectrum antibiotics in the form of meropenem 1 gm slowly intravenous every 8 hours in this period (four days). Feeding started on the second day of ventilation to all patients through a feeding pump at a rate of 70 ml insure plus (Abbot company) with 1.47 kilo-calorie/ml to supply patients with approximately 2500 kilo-calorie in 24 hours calculated by approximately 35 kilo-calorie/kg. The 5 points of bundle for VAP prevention were strictly applied to all patients: Elevation of the head of the bed 30º to 45º, daily evaluation for possible extubation, the use of endotracheal tubes with subglottic secretion drainage, oral care with oral antiseptics, initiation of safe enteral nutrition within 24-48 hours from ICU admission and ventilation.

200 patients included in our study from those who completed ventilation for 4 days and showed no improvement and still had unresolved ARDS diagnosed in our study (PO₂ /FIO₂ less than 200, bilateral parenchymatous lung infiltration on the chest X ray, ABG showed PH > 7.30), had severe sepsis which diagnosed in our study by >12 on SOFA score [1, 9], developed VAP as a complication from mechanical ventilation which diagnosed in our study by > 6 on CPIS score [10], were randomly allocated in two groups, 100 patients in each. The randomization sequence was created using Excel 2007 (Microsoft, Redmond, WA, USA) with a 1:1 allocation using random block sizes of 2 and 4 by an independent doctor. In this way, sequence generation and type of randomization can be expressed at the same time.

All patients selected underwent a percutaneous tracheostomy on the same day of selection.

Severe sepsis was documented in our study by > 12 on Sequential Organ Failure Assessment (SOFA) score illustrated in Table 2. While VAP was documented in our study by >6 on CPIS score illustrated in Table 3. Patients of group A continued on the conventional way of management mentioned previously. Bronchoscopies were done three times for patients of group B only. These bronchoscopies were done on the first, sixth and eleventh day of the studied periods. Bronchoscopies were done with the following precautions: The treating team used a flexible bronchoscopy (Olympus BF-160) adult size for this maneuver, all patients were kept sedated by both midazolam and fentanyl infusion to get sedation score (RASS-2/-3), FIO₂ was increased to 100% during the procedure, xylocaine spray 10% was used (Astra Zeneca company), two puffs were applied in each nostril before the rubber tube of the bronchoscope was introduced into the nose of those patients. Patient's head was kept elevated at 40 degrees during the procedure. The ventilator was adjusted on CMV (controlled mechanical ventilation mode) with same previous parameters mentioned before. During the procedure, four syringes of 10 ml of normal saline (NS) were injected through the bronchoscopy into the airways for washing, followed by suction of that saline in order to keep the airways clean and dry. The fluid sucked after injection of the first 10 ml NS (BAL) was sent for qualitative culture. Monitoring of patients during the procedure was done by using SPO₂, non-invasive blood pressure measurement, and electro cardiac gram.

The duration of the study was selected to be 2 weeks and the evaluation of all patients in both groups was done in three periods, at the end of the first 5 days, at the end of the second 5 days, and at the end of the last 4 days.

All patients in both groups were followed by APACHII score (acute physiological assessment and chronic health evaluation), GCS, parameters of both SOFA score and CPIS score, hemodynamic parameters (the use of inotropes and the stability of mean arterial blood pressure during the studied period) and laboratory parameters by measuring both LDH level and C-Reactive protein level.

APACHII score, GCS, core temperature (one parameter in CPIS score), and hemodynamic stability were used as indicators for general condition improvement. While controlling all parameters of both SOFA score and CPIS score used as indicators for sepsis control. Both compliance measurements and responses to recruitment maneuvers were used as an indicator for lung mechanics improvement. Recruitment maneuver is considered a clinical test of lung compliance and starts to increase the peak inspiratory pressure to 40 cm/H₂O for 40 seconds and observe the saturation (SpO2) if improved to more than 95% considered responders in our study) [12]. Both hypoxic index and parenchymatous lung infiltrate in the chest X-ray were used as indicators of ARDS improvement.

Both LDH and C-reactive protein were used as a laboratory indicator to decrease lung tissue destruction and recovery of lung tissue from both VAP and sepsis. Results of both enzymes were collected and compared among all patients in both groups in the studied period. The number of patients weaned from the ventilators in both groups was recorded and compared. Extubated patients stayed in the ICU for 24 hours for observation and were then discharged from the ICU. The number of discharged patients were recorded and compared.

Three microbiological results were collected from all patients during the studied period, at the end of the first five days, at the end of the second five days and the last one collected at the end of the studied period to confirm the bacteriological cure. Microbiological results were collected from patients of group A through qualitative sputum. While microbiological results were collected from patients of group B through BAL.

If one/all measured indicators showed no improvement in any patient in both groups during the studied period, this considered morbidity documented and compared between the two groups. A number of patients who failed to be weaned from the ventilator during the studied period, considering morbidity, were also documented and compared between the two groups. A number of patients discharged from the ICU within the studied period were recorded and compared between the two groups.

Table 4 represents the demographic data of patients in both groups and shows no significant difference between the two groups regarding age, sex, and causes of ventilation included in our study.

Table 5 compares the mortality rates in both groups all over the studied duration and shows non-significant differences between the two groups in all studied durations as (5, 5 and 4 patients) died at the end of 5,10,14 days consecutively from group A while (4, 5and 4 patients) died from group B at the same time of the study.

Table 6 compares APACH II scores of patients in both groups all over the studied duration and shows significant improvement in that score in patients of group B in all studied duration as (35,54 and 65 patients) score <15 at the end of 5, 10, and 14 days consecutively, while (9, 24 and 26 patients) in group A achieved the same score at the same time of the study.

Table 7 compares the GCS scores of patients in both groups all over the studied duration and shows significant improvement in that score in patients of group B in all studied durations as (36, 64 and 75 patients) had a score >12 at the end of 5, 10, and 14 days consecutively, while (10, 35 and 51 patients) in group A achieved the same score at the same time of the study.

Table 8 compares MAP of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (34, 60 and 74 patients) had MAP >95 mmHg at the end of 5,10,14 days consecutively, while (11, 25 and 50 patients) in group A achieved the same MAP at the same time of the study.

Table 9 compares the bilirubin levels of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied duration as (40, 67 and 72 patients) had bilirubin <1.9 mg/dL at the end of 5, 10 and 14 days consecutively, while (19, 34 and 48 patients) in group A achieved same bilirubin level at the same time of the study.

Table 10 compares the creatinine levels of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (44, 65 and 75 patients) had creatinine <1.9 mg/dL at the end of 5,10,14 days consecutively, while (20,32and 45 patients) in group A achieved same creatinine level at the same time of the study.

Table 11 compares the platelets count of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (66, 78 and 82 patients) had platelets > 100.000 at the end of 5, 10, and 14 days consecutively, while (35, 43 and 54 patients) in group A achieved same platelets count at the same time of the study.

Table 12 compares the nature of the tracheal secretion of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (55, 71 and 83 patients) had normal tracheal secretion at the end of 5, 10 and 14 days consecutively, while (19, 34 and 48 patients) in group A achieved same tracheal secretion at the same time of the study.

Table 12 compares parenchymatous lung infiltrates on the chest X-ray of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (61, 75 and 85 patients) had almost normal chest X-ray at the end of 5, 10 and 14 days consecutively, while (23, 38 and 65 patients) in group A achieved same chest X-ray at the same time of the study.

Table 12 compares the core temperature of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (50, 77 and 81 patients) had core temperature > 36.5 and < 38.4 at the end of 5, 10 and 14 days consecutively, while (25, 42 and 60 patients) in group A achieved same core temperature at the same time of the study.

Table 12 compares the total leucocytic count of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (51, 70 and 83 patients) had leucocytic count > 4000 and < 11000 thousand/mL at the end of 5,10 and14 days consecutively, while (27, 35 and 63 patients) in group A achieved same leucocytic count at the same time of the study.

Table 12 compares the hypoxic index (PaO2/FIO₂) of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied duration as (69, 76 and 83 patients) had hypoxic index >240 at the end of 5,10 and14 days consecutively, while (57, 61 and 72 patients) in group A achieved same hypoxic index at the same time of the study.

Table 12 compares the microbiological results of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (75, 80 and 86 patients) had negative qualitative BAL culture at the end of 5,10 and14 days consecutively, while (59, 62 and 70 patients) in group A had negative qualitative sputum culture at the same time of the study.

Table 13 compares lung compliance in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (32, 57 and 68 patients) had measured lung compliance of > 80 ml/ 1 cm H₂O at the end of 5, 10 and 14 days consecutively, while (14,24and 35 patients) in group A achieved same measured lung compliance at the same time of the study.

Table 14 compares the number of patients who responded to the recruitment maneuver in both groups all over the studied duration and shows a significantly higher number of patients who responded to this maneuver in group B in all studied durations as (35, 71 and 82 patients) were responders at the end of 5, 10 and 14 days consecutively, while (14, 24 and 35 patients) in group A were responders in the same time of the study.

Table 14 compares the number of patients who failed to be weaned in both groups at the end of the studied duration and shows a significantly lower number of patients in group B (only 2 patients out of 87 patients) compared to group A (17 patients out of 86 patients.

Table 14 compares the total number of patients discharged from the ICU at the end of the studied period and shows a significantly higher number of patients discharged from group B (85 patients) compared to group A (69 patients .(

Table 15 compares LDH levels of patients in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (45, 66 and 73 patients) had LDH < 200 U/L at the end of 5,10 and14 days consecutively, while (21, 33 and 40 patients) in group A achieved same LDH level at the same time of the study.

Table 16 compares C-reactive protein levels in both groups all over the studied duration and shows significant improvement in patients of group B in all studied durations as (68, 79 and 80 patients) had C-reactive protein levels <100 mg/L at the end of 5, 10 and 14 days consecutively, while (33, 39 and 49 patients) in group A achieved same C-reactive protein level at the same time of the study.

Table 17 compares the morbidity recorded in all patients between the two groups in all studied periods.

As regard, improving the general condition of the patients [ 13]:

There was a significant improvement of APACH II, GCS, and hemodynamics stability between patients in group B compared to patients in group A in all periods of the study due to more rapid and efficient control of all local lung parameters of infection measured in our study by CPIS score and more rapid and efficient control of all general parameters of severe sepsis measured in our study by SOFA scores. The pathophysiological mechanism of this finding is explained in detail later.

As regard controlling all parameters of sepsis (SOFA) score:

There was a significant improvement in the total leucocytic count, core body temperature, creatinine level, bilirubin level, and platelet count between patients in group B compared to group A in all periods of the study. This could be due to better control of systemic manifestations of both VAP and sepsis by removing the purulent discharge from the small airways. This purulent discharge might be the primary cause of sepsis. Also, this purulent discharge might be responsible for spreading a resistant bacteria to blood, causing septicemia and triggering a systemic immunological cascade of septic reactions that end by marked local and systemic tissue destruction. Moreover, patients of group B were given antibiotics according to BAL results which were considered more reliable than sputum culture. As sputum culture is associated with higher bacterial contamination than BAL, yielding of bacteria in BAL is better than sputum. All those aspects increased the success rate of antibiotic action and accelerated the bacteriological cure between patients of group B compared to group A. Also, those aspects might be the cause of a higher percentage of unsuitable antibiotics usage among patients of group A. The rapid bacteriological cure in patients of group B might be the main cause of rapid control of systemic manifestation of sepsis and septic shock. This might be the main cause of rapid recovery from a low tissue perfusion state and rapid restoration of the MAP with satisfactory tissue perfusion pressure in all organs. Therefore in those patients, there was a more rapid improvement of GCS, creatinine level, bilirubin level, and platelets count due to improvement of perfusion pressure of brain (recovery from septic encephalopathy) kidneys, liver and bone marrow, respectively. Also, there was a more rapid improvement in APACHI II score, leucocytic count and core temperature due to a more rapid improvement of global tissue perfusion and decrease/stop the release of systemic toxins and pyrogens.

As regards controlling all parameters of CPIS score, ARDS, VAP and improved lung mechanics [13]: There was significant improvement between patients in group B compared to patients in group A in oxygen saturation, hypoxic index, compliance, and response to recruitment maneuver during all periods of the study. Moreover, radiologically [13], patients of group B showed significant improvement in parenchymatous lung infiltrate on chest X-ray all over the duration of the study compared to patients of group A. This could be due to better aeration of small airways after washing and removing the purulent discharge and debris which improve lung compliance by removing the physical obstruction of pus from the small airways (as good lung aeration is one of the most important parameters of normal lung compliance). Better lung aeration also causes better lung tissue oxygenation and pulmonary vasodilation (as increased lung tissue oxygenation causes reflex pulmonary vasodilatation and increases blood supply of the lungs and thus improves local immunity of the lungs). Both improvement in lung compliance and local immunity of the lung accelerate the rapid healing from VAP and rapid control of sepsis, and thus accelerate the weaning from the ventilator.

As regards controlling laboratory markers of tissue destruction [13]:

There was a significant decrease in the markers of tissue destruction (LDH and C reactive protein) in patients of group B compared to group A. This could be due to better control of both general and local parameters of VAP, ARDS, sepsis, and/or septic shock by the same mechanism described above, which stops the cascade of sepsis and thus stops tissue destruction and rapidly reverses the state of low tissue perfusion in sepsis and accelerates the global tissue reperfusion.

As regards lowering morbidity and shortening the duration of ICU stay [13]:

The number of patients who showed no improvement of one/or all parameters of both CPIS score and SOFA score was significantly higher in group A compared to group B, as illustrated in Table 17, and this could be explained by the lower number of weaned patients from the ventilator in group A compared to group B. And prolonged ventilation is the main cause of morbidity in all patients in the critical care; moreover, patients of group B showed shorter duration of ICU stay compared to group A due to higher number of weaned patients from the ventilator. No significant difference was found in the mortality rate between both groups.

Our study supports those studies done on this aspect as Fagon et al., in 2000 [14] demonstrated lower short-term mortality and decreased Sequential Organ Failure Assessment Scores at three days (6.1 – 4.0 vs. 7.0 – 4.3 points; p = 0.33) and seven days (4.9 – 4.1 vs. 5.8 – 4.4 points; p = 0.04) for their group that had a diagnosis via BAL versus a non-invasive technique. This finding suggests that bronchoscopy has some short-term therapeutic benefits. Michetti et al., in 2012 [15] did a study on 137 patients with VAP due to lung contusion following chest trauma; they compared BAL-based microbiological results versus endotracheal sputum-based microbiological results, and they found improvement in signs of both VAP and lung contusion and this was associated with low overall mortality and VAP-attributable mortality. Christopher A et al., in 2014 [16] did a trial on 360 patients with VAP, and they compared the invasive diagnosis of VAP by bronchoscopy with the noninvasive traditional way of diagnosis. They found that diagnostic bronchoscopy was associated with a shorter length of stay and shorter duration of antibiotics. Qiao, Zhihao MM et al., in 2018 [17] did a study on 107 patients with respiratory failure patients due to acute exacerbation of COPD and compared daily suction of sputum by bronchoscope with general suction by the traditional way. They found that the bronchoscopic sputum suction group showed the earlier appearance of normal sputum color, lesser viscosity, scanty in amount, shorter time of invasive ventilation, total time of ventilation and hospital stay, lower reintubation rate, and fatality rate, and higher weaning success rate than the general sputum suction group (all P < .05). On the other hand, Sanchez-Nieto et al., in 1998 [18] and Ruiz et al., in 2000 [19] found no statistical difference between the culture results obtained from BAL and those obtained from sputum aspirated from the endotracheal tubes in both short-term and long-term results of VAP as regards the duration of antibiotics, duration of ventilation, duration of ICU and mortality rates. Shorr et al., in 2005 [20] and Berton et al., in 2012 [21] did two meta-analyses (including a Cochrane review) and found no statistical difference between microbiological results obtained from BAL and those obtained from sputum on the duration of antibiotics, duration of ventilation and duration of ICU stay. Another randomized trial done on 740 patients who had VAP in 28 ICUs in Canada and the United States published by the Canadian Critical Care in 2006 in this study compared bacteriological results taken by BAL versus sputum taken from the endotracheal tube and found no significant difference as regards the duration of antibiotics, duration of ventilation, APACHII score and mortality rate between the bronchoalveolar-lavage group and the endotracheal-aspiration group [22].

Von Essen et al. in 1991 [23], Lawrence et al. in 1996 [24], Sato et al. in1998 [25], and Takeshi et al. in 2001 [26] did trials on the use of bronchoscopy in pneumonia and had significant signs of deterioration of their patients after bronchoscopy and they did not recommend its use in pneumonia.

From all previous studies, we can divide the authors' opinion on this aspect into three main categories, a group of authors proved that bronchoscopy in VAP is very beneficial and significantly improves both short-term outcomes (shortening the duration of antibiotics therapy, ventilation and ICU stay) and long-term outcomes, (morbidity and mortality rate of VAP) another group of author found no significant difference between the use of bronchoscopy in both diagnostic and/or therapeutic management of VAP, following the traditional way of management of VAP. The last group of authors found significant deterioration signs in patients who had bronchoscopy with VAP, which could be due to the breaking of the localizing mechanism of the body to isolate the inflamed area of the lung and spread the infection to the adjacent lung segments, facilitate the systemic spread of the infection to the blood stream, increase the ventilation perfusion mismatch through obstructing the small airway with the wash during bronchoscopy, and aggravate the systemic cascade of sepsis through flooding the systemic circulation with purulent discharge from the inflamed lung areas. We found bronchoscopy very beneficial in our center when used in VAP and septic patients and had significant results proving improvement in both short- and long-term outcomes of both VAP and sepsis except for mortality rate, which showed no significant difference between both groups.

There were many limitation points in our study, including small sample size, predominantly male sex in our sample, exclusion of pediatric age group, using only SOFA score for sepsis, not all parameters of sepsis, selecting only VAP, type of health care acquired pneumonia and not all types of health care acquired pneumonia, and only 4 lung diseases (traumatic lung contusion, COPD with bronchitis, Bronchiectasis and aspiration pneumonia) were used in our sample size which are widespread to cause ARDS in our locality. However, still, our study is considered a unique study that follows the effect of bronchoscopy in both severe sepses with VAP as this combination has never been recorded before. We recommend more research work on the role of bronchoscopy in all types of pneumonia, not only VAP, and emphasise a regimen for its use in this aspect. Moreover, the same study is needed in pediatric patients as our study was done only in the adult age group.

The use of bronchoscopy can improve general conditions, control all signs of severe sepsis and VAP, improve lung mechanics, improve ARDS, accelerate weaning from the ventilator, and shorten the ICU stay but has no effect on mortality rate in those patients with severe sepsis with ARDS and complicated by VAP from prolonged ventilation.