Introduction
Mechanical ventilation is an essential component during intensive care management of critically ill patients. Many patients require ventilatory support in the postoperative period following extensive and protracted surgery. Patients with severe head injury also require mechanical ventilation, at times for prolonged period to maintain systemic oxygenation and intracranial pressure within acceptable range. Servo controlled ventilator is ideal in these situations. However, the major constraints are its exorhitant cost and availability in a busy intensive care unit like ours. Also in the event of break down, rectification might take a long time further adding to the cost of ventilator. To overcome all these above problems, a search for indigenous ventilator is still continuing. Previously we had evaluated the performance of early version of Medisys system-l both in the laboratory, and in clinical set up. 'The same ventilator was evaluated further for providing prolonged ventitatory support.' Recently, a newer version of the indigenous ventilator [Mediveni Excel. Medisys] has been launched to provide mechanical ventilation in the ICU set up. Thus the present study was undertaken to evaluate the adequacy and efficacy of this ventilator by using a novel method of comparison of its performance with that of a servo-controlled ventilator [Horus Taema, Francel in the clinical setting.

Materials and methods
The indigenous ventilator used in the present study [Medivent Excel, Medisys] (Fig. 1) is an electrically driven, time cycled, volume preset, pressure limited, constant flow generator. Compressed gas source is not required for driving the ventilator. It provides three modes of ventilation: controlled mode (CMV), assist

control mode [ACVI and synchronized intermittent mandatory ventilation mode [SIMV]. Oxygen flow of 2-4 Lmin-1 is required so as to deliver 40% O2 [F1O2 0.4.]. It flows through a Venturi device and gets collected in a reservoir bag of 2 L capacity, from which the bellows. extract the set amount of volume and delivers it to the patient through the circuit. With the help of Venturi device attached to the equipment, oxygen can be provided at three different concentrations - 40, 50 and 100%. The front panel has an airway pressure dial and electronic display of adjustable variables like respiratory rate, tidal volume, inspiralory hold. mode of ventilation, 1: E ratio and low pressure alarm. Safety pop-up valve device can be set up at 20,40, and 60 cm H2O) The safety alarm system was checked using test lung (reservoir bag) and was within the specification range quoted by the manufacturer. All the parameters can be changed by the use of a single knob. The ranges of parameters are provided in table I. The ventilator allows the patient to breathe freely through the circuit between the mandatory breaths.

After obtaining approval from Institutional Ethics Committee and informed consent from next-of-kin, 32 critically ill adult patients of either sex, admitted to our neurointensive care unit were recruited into the study. Patients with age less than 18 years old, patients with pulmonary diseases and high cervical spinal cord lesions were excluded from the study.

All patients were sedated and ventilated in the controlled mode [CMV] during the study period. During this period, each patient was ventilated for 6 hours each using indigenous and servo-controlled [Horus, Taema , France] ventilators consecutively. First 16 patients were initially ventilated with indigenous ventilator for 6 hours and then changed over to servo controlled ventilator for another 6 hours, while the reverse was carried out for the rest 16 patients. Each ventilator was set to deliver a tidal volume of 10 mlkg-1 , F1O2, 0.4, respiratory rate 10 min ' and inspiratory hold was adjusted to maintain 1: E ratio of 1:2.

During the study period, monitoring of physiologic variables included heart rate [HR], mean arterial blood pressure [MAP, noninvasive], side stream capnometry and pulse oximetry [AS/5. Datex Ohmeda, Helsinki, Finland ] and recordings were made at half hourly intervals. Simultaneously F1O2 delivered tidal volume, total respiratory rate, airway pressure, compliance, and resistance were monitored using the spirometry module [ D-Iite, AS/5, Datex Ohmeda, Helsinki, Finland ]. Arterial blood gas analysis was performed at hourly intervals. The neurological status of the patients were assessed with the help of Glasgow coma scale and pupillary signs. At the end of study period [I2 hours], patients were either connected back to servo-controlled ventilator or continued to receive ventilatory support from indigenous ventilator.

Statistical analysis was performed using multiple paired t-test. A p-value of less than 0.05 was considered significant.

Results
Out of the 32 patients included in the study, 17 were males and the rest females. The mean [±SD] age was 50 [±8] years and the mean weight was 65[±5] kg. The conditions for which the patients received ventilatory support are depicted in table 2.

The respiratory parameters during ventilation with each type of ventilator were compared with spirometry values and are represented in table 3 and 4. The tidal volume recordings in spirometry was slightly less as compared to the set value in the Medivent, but not significant. in case of Taema ventilator, recorded value in spirometry was slightly greater but insignificant. However, when tidal volumes of both Medivent and

Taema ventilators were analyzed, there was no significant difference. There were no significant differences in F1O2, and peak airway pressure delivered by the indigenous ventilator and the standard ICU ventilator with the recordings made in the spirometry. Ventilatory parameters [ETCO2, PaCO2] oxygenation parameters [SpO2, PaO2] and haemodynamic parameters [heart rate, blood pressure] were also similar during ventilation with both types of ventilators [table 5].

Discussion
The present study has shown that the set respiratory parameters [F1O2 tidal volume, respiratory rate] of both the ventilators, Taema and Medivent, correlate very well with the spirometric values recorded with the help of Datex Ohmeda AS/5 monitor. Comparative evaluation of the expired values in the spirometry during ventilation with Taema and Medivent ventilators has also shown very good correlation. The blood gases and haemodynamic parameters are also comparable during ventilation with both ventilators. Hence in normal lung, Medivent, an indigenous ventilator provided adequate controlled ventilation, which was comparable to that provided by Taema, a servo controlled sophisticated ventilator.

Evaluation of a mechanical ventilator ideally has to be performed in different stages. First stage involves the verification of delivery of tidal volume, respiratory rate and airway pressure with the help of Wright's respirometer and pressure manometer. Second stage includes the study of flow pattern and airway pressure recording with the help of artificial lung model in normal as well as in abnormal conditions [for example: reduced compliance, increased resistance]. Once all these are found satisfactory, then evaluation has to be undertaken in clinical setting in patients with normal and diseased lungs. 3-5 All the above processes need the help of good respiratory laboratory and are very tedious and time consuming. Good rapport with respiratory physician is also imperative. Literature search revealed only few studies, which actually describe ventilator performance assessment, that too using lung models. As we do not have such provision in our set up, we embarked on a novel method.

We have utilized our critical care monitor [Datex AS/5] to conduct the present study, as it has spirometry facilities. The spirometry provides both inspiratory and expiratory measurements of tidal volume, respiratory rate, airway pressure, compliance, resistance, and pressure volume loop for every breadth. These values are stored in the monitor and can be retrieved. Thus we feel that there is no other better way than comparing the set respiratory parameters of the ventilator with that of spirometry recordings. We used the Taema ventilator for comparison, as it is servo controlled, sophisticated, and also has display screen which shows expiratory tidal volume, respiratory rate, airway pressure, pressure volume loop, and F1O2. These values were similar to spirometry recordings, thereby proving (he adequate functioning of spirometer and multiple gas analyzer.

Since postoperative neurosurgical and head injury patients form the major patient load in any neurointensive care unit, we included these patients in our study. These patients usually have normal lungs and they require mechanical ventilation due to poor Glasgow Coma Scale, reduced respiratory drive, hypothermia, cerebral edema, seizures, or residual anaesthetic effects rather than due to pulmonary diseases or sepsis. Controlling arterial carbon dioxide level and maintaining, adequate oxygenation are important during ventilating a neurologically ill patient. Volume preset ventilators are most efficacious in achieving these goals without causing significant detrimental effects. They also help to prevent atelectasis task and hypoxemia. Hence the indigenous, volume preset ventilator was tested for its performance so that it can be used in intensive care unit to accomplish the above mentioned goals.

The present study is not without its limitations. Recordings were done for only 6 hours each during ventilation with either type of ventilator. But, critically ill patients might require longer periods of ventilation. As we collected the recordings at half hourly intervals for 6 hours, we feel that the data is sufficient enough to be extrapolated to longer periods of ventilation. The study was conducted in haemodynamically stable patients with normal lungs. Therefore, its performance in abnormal lung conditions needs to be evaluated. Also, the performance of ventilator in pediatric patients requires further study.

Although the indigenous ventilator has been found to provide adequate ventilation in patients with normal lungs, the following recommendations have to be implemented to make it more versatile.

1. Venturi device needs to be modified to provide varying concentration of oxygen to the patient.
2. Provision of humidifier.
3. Provision of positive end-expiratory pressure [PEEP] device.
4. Triggering must be variable. Addition of PEEP to this indigenous ventilator with fixed triggering might increase the workload of the patient.
5. Display of expiratory respiratory parameters.
6. Provision of other modes of ventilator like pressure support mode, which is useful for weaning Purposes-
7. Provision of battery backup.
   
   In conclusion, the present study confirms the accuracy of delivery of indigenous ventilator with no associated significant deterioration in respiratory and haemodynamic variables. Further study is being designed assess the performance of this ventilator in weaning patients from ventilatory support using SIMV mode in the intensive care unit.

References

1. Kaul HL, Dash HH, Arora SK. Code GR. Evaluation of an indigenous anaesthesia-ventilator. Indian J Anaesth 1984; 32: 369.
2. Mohanty B. Bhutra S, Chauhan RS, Dash HH.. is Medisys system I ventilator effective in providing IPPV in severe head trauma patients. Iindian J Anaesth 2001;. 45: 24-29.
3. McCluskey A, Gwinutt CL. Evaluation of the Pneupac Ventipac portable ventilator: Comparison of performance in a mechanical lung and anaesthetized patients. Br J Anaesth 1995; 75:645-50.
4. McGluskey A, Hardy L, Haslett R, Bowles B, Kishen R. Evaluation of the Pneupac Ventipac portable ventilator in & critically patients. Anaesthesia 2001; 56: 1073-81.
5. Brill S, Gurman GM, Brill G. Evaluation of the Versa Med portable ventilator: clinical trials. Eu J Anaestliesiol 2000; 17: 737-43.