|Year : 2006 | Volume
| Issue : 2 | Page : 70-74
Evaluation of the predictors for duration of mechanical ventilation in respiratory intensive care unit
HS Hira, Amit Mittal
Department of Medicine, Pulmonary Medicine and Sleep Center, Maulana Azad Medical College and Lok Nayak Hospital, New Delhi - 110 002., India
H S Hira
House 74, Block 21, Lodi Colony, New Delhi - 110 003.
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Mechanical ventilation (MV) is a life saving modality. Till no method is evolved to predict the duration of this treatment. This study is undertaken to evaluate the clinical and laboratory parameters at initiation and 48 hour of the patient being treated by mechanical ventilation; and correlate these parameters with the duration of MV required.
Methods: It was prospective observational study conducted in the respiratory intensive care unit (RICU) of a tertiary referral and teaching hospital. Thirty consecutive patients suffering from various medical diseases requiring MV were the participants. Clinical, laboratory, ventilatory parameters and six severity scores viz.; GCS, APACHE-II, SAPS, OSF, ALI/ARDS and Sepsis/Septic shock criteria of each patient, both at initiation and at 48 hour of MV were recorded. Correlation between various severity scores at initiation and at 48 hour of initiation with duration of mechanical ventilation was computed.
Results: Based on the duration of mechanical ventilation required, the patients were divided in two groups. Those requiring mechanical ventilation <7 days and >7 days were designated Group I and Group II respectively. Three out of 6 severity scores (GCS, APACHE-II, SAPS) recorded at 48 hour of MV showed statistically significant difference between the both groups with p value of 0.010, 0.009 and 0.006 respectively. Further stepwise logistic regression analysis showed that SAPS score at 48 hour of initiation was the best predictor of duration of MV. It was found that a cut off value of 15 for SAPS score at 48 hour might predict the duration of MV i.e. <7days or >7 days. Based on linear regression analysis a simple equation was formulated by putting the SAPS value at 48 hour, the value of Y was computed. If MV required was less than 7 days, the value of Y was more than - 0.172 and if need of mechanical ventilation was more than 7 days, Y was less than -0.1720.
Conclusions: For patients requiring MV the clinical, laboratory, ventilatory parameters and severity scores at 48 hour of initiation of MV were more significantly correlated to duration of MV as compared to those at initiation. So the parameters and severity score at 48 hour may help to anticipate the approximate duration of MV. Further a cut off value of 15 for SAPS score at 48 hour helped to predict the duration of mechanical ventilation. The calculation of Y from evolved equation is presumed to predict the duration of requirement of MV.
|How to cite this article:|
Hira H S, Mittal A. Evaluation of the predictors for duration of mechanical ventilation in respiratory intensive care unit. Lung India 2006;23:70-4
|How to cite this URL:|
Hira H S, Mittal A. Evaluation of the predictors for duration of mechanical ventilation in respiratory intensive care unit. Lung India [serial online] 2006 [cited 2020 Oct 25];23:70-4. Available from: https://www.lungindia.com/text.asp?2006/23/2/70/44412
| Introduction|| |
Mechanical ventilation (MV) is imperative in many forms of acute respiratory failure. Knowledge about duration of MV for critically ill patient helps in several ways particularly if it is predictable beforehand. If the approximate duration of MV required is known in the beginning, clinician shall be able to foresee the timing of weaning and also may perform early tracheotomy to avoid complications.
How long critically ill patient needs MV varies from one patient to the other. There are no existing parameters to predict it accurately. It depends on severity of illness, which in turn is determined by several factors viz: age, underlying disease process, pre/co-existing morbid illness, and organs involved (single/multiple; involvement of vital organs), physiological derangements and presence/absence of sepsis. Multiple scoring systems have been evolved to evaluate the severity ,,,,, viz. Glasgow Coma Scale (GCS), Acute Physiology, Age and Chronic Health Evaluation-II (APACHE-II), Simplified Acute Physiology Score (SAPS), Organ System Failure (OSF), Acute Lung Injury and Acute Respiratory Distress Syndrome Criteria (ALI/ARDS criteria) and Sepsis and Septic Shock Criteria (S/SS criteria).
In general more severe the illness, more prolonged is the requirement for MV, but this is not true in extreme critical situation, when the patient dies within a short period leaving little scope for MV. The parameters noted at initiation of MV did not take into consideration the importance of early developed sepsis and the improvement in various physiological parameters of the patient after the initiation of MV. Therefore a further evaluation of these parameters at 48 hour of initiation is required to predict the approximate duration of MV correctly.
| Material and Methods|| |
The study was conducted in the respiratory intensive care unit (RICU) of a tertiary referral and teaching hospital. Thirty patients suffering from various medical diseases requiring MV over a year in RICU were included. Academic board of the institution approved the protocol.
The general criteria for initiation of MV were followed and were also modified according to the clinical status of the patients. These criteria included: respiratory rate > 35/min, visible excessive work of breathing, clinical evidence of respiratory fatigue, tidal volume <5 ml/kg, minute volume >10-12 L/min over a prolonged period, negative inspiratory pressure (NIP) <20 cmH 2 O, PaO 2 <60 mmHg on nasal oxygen (FiO 2 0.6) at 6-8 L/min, and/or PaCO 2 >55 mmHg, P (A-a) 0 2 more than 300-350 mmHg on FiO 2 0.6. Patients requiring MV for cervical trauma and those with known subglottic or tracheal abnormality were excluded. The T-Bird AVS III ventilator (15322C, 2001) with MR410 respiration humidifier were used for mechanical ventilation (MV).
On initiation of mechanical ventilation, clinical details of the patients including age, sex, presenting disease, background illness, temporal profile, timing of intubation and timing of tracheotomy (if any) were recorded. Laboratory profile of each patient e.g. complete hemogram including hematocrit, kidney function tests and serum electrolytes, liver function tests with serum proteins, prothrombin time, blood sugar, arterial blood gas analysis (ABG), chest radiograph and electrocardiograph (ECG), blood/ urine/sputum culture and sensitivity (if any) and six severity scores viz. Glasgow Coma Scale, Acute Physiology, Age and Chronic Health Evaluation-II, Simplified Acute Physiology Score, Organ System Failure, Acute Lung Injury and Acute Respiratory Distress Syndrome Criteria and Sepsis and Septic Shock Criteria were recorded. All patents were evaluated again by recording the same severity scores at 48 hour after initiation of MV.
For statistical analysis patients were grouped into two, depending on duration of MV needed, those who required mechanical ventilation for <7 days (Group I) and patients who required mechanical ventilation for >7 days (Group II). The students T- test and Fischer's exact test were applied to assess the differences between both groups for quantitative and qualitative variables respectively. Stepwise logistic regression analysis was performed for those scores that are significantly different between two groups.
| Results|| |
Out of 30 patients, 19 patients required mechanical ventilation (MV) for <7 days (Group I) and 11 patients required MV for >7 days (Group II). Mean age of the patients included in study was 53+19 years. The mean age of the patients was 51.05+20.2 years in Group I and 56.8+16.9 years in Group II, which was not statistically significant. There were 22 male patients and 8 female patients, with a sex ratio of 2.8:1. There was no significant difference in sex distribution in both groups. Average duration of mechanical ventilation support and RICU stay was 16.1+26.3 and 19.4+26.6 days respectively. However average duration of mechanical ventilation and RICU stay was 3.63+2 and 6.21+3.45 days in group I and 37.72+34.44 and 42.18+33.75 days in group II.
The distribution of basic disease groups and presence of past illness was also same between two groups. The neurological disorders required maximum duration of MV (25.4+49.01 days) while those with ARDS required minimum duration of mechanical ventilation i.e. 2.5+0.71 days [Table 1]. Of the total 30 patients included in this study, 14 (46%) survived and 16 (54%) died, survival in group I and group II was 47.4% (9 patients) and 45.5% (5 patients) respectively.
Eight (26%) patients underwent tracheotomy. Its average duration from starting of MV was 8+11.4 days, with minimum interval being one day (for tetanus patients) and it was 32 days (patient of COPD with acute respiratory failure patient). Average duration of mechanical ventilation after tracheotomy was 38.1 + 38.9 days compared to 8.1 + 14 days in patients without tracheotomy; this seemed to be significant but the p value was 0.067 (not significant). Majority of the patients underwent tracheotomy were in group II.
Most of the clinical, laboratory and ventilatory parameters at initiation and 48 hour did not show any significant difference between group I and group II. None of the severity scores (GCS, APACHE-II, SAPS, OSF, AWARDS criteria, S/SS criteria) at initiation showed statistically significant difference between the two groups. At 48 hour of initiation, 3 of 6 severity scores viz. GCS, APACHE-II and SAPS showed statistically significant difference between the both groups with p value of 0.010, 0.009 and 0.006 respectively [Table 2]. These three scores were put in stepwise logistic regression analysis with duration of MV as dependent variable and other three scores as independent variables. It was seen that at step two of regression analysis, only SAPS at 48 hour was significantly correlated with duration of MV required with p value of 0.006. Further SAPS values were analyzed using receiver operating characteristic (roc) curve to find a value of SAPS which could signal whether patient remained on MV for <7 days or >7 days. It was found that with the SAPS value of 15, the specificity took over the sensitivity; if the calculated value of SAPS at 48 hour was <15, the duration of MV was less <7 days and when the SAPS value was > 15, the duration of mechanical ventilation was >7 days. Linear regression analysis was done keeping SAPS at 48 hour as independent variable and duration of mechanical ventilation as dependent variable. Linear regression equation was formulated as:
Y = -6.292 + 0.408(SAPS at 48 hour)
After putting the SAPS in the equation if the value of Y was more than -0.172, the patient required mechanical ventilation for <7 days. Patient needed mechanical ventilation for >7 days when Y was computed to be less than -0.172.
| Discussion|| |
Mechanical ventilation (MV) is the life saving supporting device for failing respiratory functions. If the duration of MV is known in advance at the time of its initiation, it would have helped to decide about the timing of tracheotomy and the expected timing of weaning.
There was no uniform consensus regarding the point of time beyond which ventilation should be considered as prolonged or beyond which complications of endotracheal tube (ETT) outweigh its benefits and benefits of tracheotomy outweigh its risk. , A conventional approach was to continue ETT when predicted duration of MV is likely to be 10 days or less and to perform tracheotomy when predicted duration of MV was to be more than 10 days. Some studies advocated that when anticipation was not feasible, daily assessment was required and ETT should be converted to tracheotomy if requirement of MV was prolonged beyond 10-14 days, unless there was imminent possibility of weaning or there were other hindering factors e.g. uncorrected coagulopathy. ,,,
Most of the studies on prediction of duration of mechanical ventilation tried to assess the expected duration of mechanical ventilation on basis of various severity score calculated on day 1 of mechanical ventilation. ,, One limitation of latter studies was that severity scores were calculated during the first 24 hour of initiation of mechanical ventilation and so the fresh evolution of complications like early developed sepsis, which could alter the outcome later, in term of further need for MV were not considered. However in present study, the severity score were calculated both at initiation and 48 hour of initiation of mechanical ventilation to take into consideration evolved complications.
Further with the introduction of MV in the patient's management there was some early improvement in physiological, laboratory and ABG parameters of the patients and the changes in these parameters were studied at 48 hour of initiation. Even small improvement in these parameters altered the severity score significantly, when scores of 48 hour were compared with those at initiation. This improvement in severity scores at 48 hour might help to predict the approximate duration of MV. So a repeat evaluation of all the clinical, laboratory and ventilatory parameters at 48 hour of initiation of MV was undertaken, and a calculation of severity scores at 48 hour was of advantages in the prediction of expected duration of MV.
In the present study, age and sex could not predict the duration of MV. In addition the disease of the patient or the cause of initiating MV and the presence of co morbid illness also didn't correlate with the expected duration of MV. A significant correlation between duration of MV and its indication was observed.  But in latter study, predominantly surgical patients were included. The duration of MV did not show any profound effect on the survival rate in the present study. Similar observation was also made earlier also. 
In the present study the mean timing of tracheotomy was 8+11 days and the mean duration of MV in these patients were comparatively longer than those managed on endotracheal tube.
In this study most of the clinical, laboratory and ventilatory parameters didn't show any significant difference between two groups both at the beginning and 48 hour of initiation of mechanical ventilation. However there was much improvement observed in these parameters at 48 hour as compared to those of zero hour. It was more so in group I as compared to group II.
The severity scores didn't show any significant difference between two groups when calculated at the initiation of mechanical ventilation. In contrast to present study, one earlier study found Lung injury score (LIS), Sepsis score (SS) and Organ system failure (OSF) scores to be significantly different between two groups (<14days vs. >14 days of MV); the scores had been calculated on day 1 of mechanical ventilation.  However one of the possible explanations for discrepancy of results between that and the present study could be that most of the patients included by them were surgical patients and latter didn't had high LIS, SS or OSF score and required MV for very short duration, a mean of 2+0.5 days.
However in present study, at 48 hour of initiation of mechanical ventilation GCS, APACHE-II and SAPS were significantly different between two groups. Thus the severity score at 48 hour correlate much more with the duration of MV as compared to the severity scores at zero hour. This observation was made in present study as none of the previous studies ,, had calculated the severity scores at 48 hour of initiation of MV. Others had found Lung injury score (LIS) evaluated on day 7, to be significantly different in two groups.  Only ARDS patients were included that study that could justify this predictability of LIS.
In present study, it was found that SAPS at 48 hour was the best independent predictor of duration of MV with a cut of value of 15 which predict that if SAPS at 48 hour was less than 15, patient needed MV for <7 days and if SAPS at 48 hour was more than 15, patient required MV for >7 days. An equation [Y = - 6.292 + 0.408 (SAPS at 48 hour)] was formulated to predict the duration of mechanical ventilation. The calculation of Y was considered to be rewarding. When it was more than -0.172, MV requirement was for <7 days. Patient remained on mechanical ventilation for >7 days in case it was less than -0.172.
It was concluded that the severity scores at 48 hour of initiation of mechanical ventilation were better predictors of duration of mechanical ventilation compared to the severity score at initiation of MV. Of all the scoring systems, SAPS at 48 hour was the best predictor of duration of MV. The calculation of Y from evolved equation might help to predict the duration of requirement of MV in a particular patient and thus aid to plan the further management strategies. For patients requiring prolonged MV, early tracheotomy can be planned, and it may provide an idea about the success of weaning in a particular patient.
| References|| |
|1.||Teasdale G, Jennet B. Assessment of coma and impaired consciousness: A practical scale. Lancet 1974,2.- 81-84. |
|2.||Knaus WA, Draper EA, Wager DP. APACHE - II: A severity of disease classification system. Critical Care Med 1985,13.- 818-29. |
|3.||Le Gall JR, Loirat P, Alperovitch A. A simplified acute physiologyscores for ICU patients. Critical Care Med 1984; 12: 975-77. |
|4.||Knaus WA, Draper EA, Wagner DP. Prognosis in acute organ system failure. Ann Surg 1985,202: 685-93. |
|5.||Gordon R. Bernard, Antonio A, Kenneth L et al. The American-European consensus conference on ARDS. Am I Respir Crit Care Med 1994; 149: 818-24. |
|6.||Roger C. Bone, Robert A. Balk, Frank B. Cerra et al. The ACCP/SCCM consensus conference on definition for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101: 1644-55. |
|7.||Whited RE. A prospective study of laryngotracheal squeals in long-term intubation. Laryngoscope 1984,94.- 367-77. |
|8.||HeffnerJE. Tracheal intubation in mechanically ventilated patients. Clinics Chest Med 1988,24.- 23-35. |
|9.||Heffner IF, Zamora CA. Clinical predictors of prolonged translaryngeal intubation in patients with the adult respiratory distress syndrome. Chest 1990; 97: 447-52. |
|10.||Heffner IF. Timing of tracheotomy in mechanically ventilated patients. Am Rev Res Dis 1993; 147: 768-71. |
|11.||Dunham CM, La Monica C. Prolonged tracheal intubation in the trauma patient. I Trauma 1984,24.- 120-24. |
|12.||Troche G, Moine P. Is the duration of mechanical ventilation predictable? Chest 1997,;112: 745-51. |
|13.||Kaplan J, WemmllerE. A method of predicting the length of intubation in trauma-induced respiratory insufficiency Laryngoscope 1983; 92: 1374-76. |
[Table 1], [Table 2]