Pediatric Pulrnonology 17:155-160 (1994)

High Frequency Jet Ventilation: lntraoperative Application in Infants

Jay S. Greenspan, MID’, Deborah A. Davis, M D ~ , Pierantonio Russo, M D ~ ,

Michael J. Antunes, M D ~ , Alan R. Spitzer, M D ~ , and Marla R. Wolfson, pho4

Summary. The potential advantages of the intraoperative use of high frequency jet ventilation (HFJV) when compared with conventional ventilation (CV) include the maintenance of adequate gas exchange and lung function with a relatively motionless surgical field. To determine the pulmonary response to HFJV ventilation in infants during cardiac surgery, we evaluated lung function in nine infants supported with CV and HFJV during a Blalock-Taussig shunt procedure. Infants were randomized to each mode of ventilation with inspiratory and expiratory pressures and Fi, held constant. Heart rate, blood pressure, arterial blood gases, pulmonary mechanics (tung compliance and resistance), and functional residual capacity (FRC) were compared after 10 minutes of stabilization of each ventilation mode, with the infants in the thoracotomy position and the surgical field adequately exposed. Pulmonary mechanics were measured using esoph- ageal manometry and pneumotachography, and FRC by helium dilution. There was no difference in vital signs, pulmonary mechanics, FRC, or Pa, on HFJV ventilation when compared with CV. Arterial Pa,, was lower with a lower mean a indy pressure on HFJV when compared with CV. The surgicafieam subjectively observed a diminished need for lung manipulation and improved ease of access to the surgical field with HFJV. These results indicate that the use of HFJV during closed-heart cardiac surgical procedures in infants provides similar cardiopulmonary stability and some potentially important clinical benefits when compared with CV. Pediatr Pulmonol. 1994; 17:155-160. 0 1994 Wiley-Liss, Inc.

Key words: Closed-heart surgery; blood gases; lung compliance and resistance; func- tional residual capacity.

INTRODUCTION

The intraoperative management of infants undergoing cardiac surgery is often complicated by pulmonary insta- bility from preoperative and intraoperative changes in lung function. ’ Further difficulties arise from attempts to expose adequately the small surgical field with lung re- traction and placement in the thoracotomy position, with subsequent atelectasis and dysfunction of affected lung units. * This necessitates high pulmonary inflation pres- sures with mechanical ventilation to achieve adequate gas exchange. However, elevation of airway pressure may disturb cardiovascular dynamics, especially in infants with little cardiovascular reserve, thereby further com- promising gas exchange.

High frequency jet ventilation (HFJV) utilizes fast res- piratory rates (greater than 150 breathdmin) and small tidal volumes (2-4 mL/kg) to maintain adequate minute ventilation and gas exchange.334 Efficacy in various neo- natal and pediatric pulmonary diseases has been demon- strated, with adequate ventilation achieved at relatively lower mean airway pressure^.^-^ The advantages of small-tidal volume ventilation with HFJV has also been utilized during airway surgery in adults to maintain gas exchange in a relatively motionless surgical field.’ In this 0 1994 Wiley-Liss, Inc.

regard, the intraoperative use of HFJV has been shown to be beneficial in patients with normal lung function.

The intraoperative pulmonary management of infants during cardiac surgery could be utilized to treat pulmo- nary abnormalities by reducing lung movement and eas- ing access to the surgical field. We hypothesized that the use of HFJV during neonatal cardiac surgery could main- tain lung volume, pulmonary mechanics, and gas ex- change with reduced lung movement when compared with conventional ventilation (CV). To test this hypothe- sis, infants undergoing a Blalock-Taussig shunting pro- cedure were evaluated intraoperatively on CV and HFJV.

From the Department of Pediatrics, Thomas Jefferson University School of Medicine’ and the Departments of Cardiothoracic Surgery,’ Ane~thesia ,~ and Phy~iology,~ Temple University School of Medicine, St. Christopher’s Hospital for Children, Philadelphia, Pennsylvania.

Received June 15, 1993; (revision) accepted for publication August 26, 1993.

Address correspondence and reprint requests to Dr. J.S. Greenspan, Division of Neonatology, Jefferson Medical College, 1025 Walnut St. , Room 700, Philadelphia, PA 19107.

156 Greenspan et al.

MATERIALS AND METHODS Patients

Nine infants born at term and less than 3 months old (mean postnatal age, 5 ? 4 SD weeks; study weight, 3 . 2 k 1 SD kg), with cyanotic congential heart disease requiring Blalock-Taussig shunting, were enrolled in this study after informed parental consent was obtained. All were free of primary pulmonary disease or intercurrent illness and all required mechanical ventilatory support at low respiratory rates preoperatively for presumed pros- taglandin El-induced apnea.

Operative Care The study population underwent a modified Blalock-

Taussig shunt procedure (left to right cardiac shunt) uti- lizing an artificial graft (W. L. Gore and Associates, Elkton, MD). For the procedure the infants had oral- tracheal intubation with an appropriately sized triple lu- men catheter (Mallinckrodt Inc., Argyle, NY), which allows for both CV and HFJV and continuous distal tra- cheal pressure monitoring. The infants were anesthetized with a fentanyl infusion and received skeletal muscle paralysis with pancuronium bromide via standard proto- col. All had standard monitoring with electrocardiogram, in-line blood pressure recording, and transcutaneous ox- yhemoglobin saturation. Arterial blood gas measure- ments were obtained via the indwelling catheter. The mean airway pressure (MAP) in the distal trachea was measured by the pressure transducer on the HFJV venti- lator.

Measurement of Lung Mechanics Lung compliance (C,) and resistance (RL) were deter-

mined with the infant in the supine and the head in the neutral position. As previously described, simultaneous signals of air flow and transpulmonary pressure were related to a software program for data analysis (PEDS, PTI Inc., Jeru~alem).~ A water-filled catheter was placed orally into the distal esophagus and was attached to a differential pressure transducer (model P7D Celesco Transducer Products, Inc., Canoga Park, CA). The cath- eter position was checked by observing the on-line moni- tored pressure tracing. The transpulmonary pressure change was measured as the difference between the air- way and the esophageal pressure. Air flow was measured with a heated pneumotachometer (Fleisch model 00, OEM Medical, Richmond, VA) and a differential pres- sure transducer (model MP45, Validyne Engineering Corp., Northridge, CA). This device was attached to the endotracheal tube with a low-volume adapter in all sub- jects (Vital Signs, Totowa, NJ). A tube from the side port of this adpater was attached to the differential pressure transducer to measure airway pressure. The resistance and the dead space of this assembly are 13.2 cmH,O/L/s

and 1.7 mL, respectively. When the infant was on CV, mechanical tidal volume breaths were analyzed. When on HFJV, sigh breaths were analyzed. Pressure and flow signals were sampled during 60 seconds. These pressure and flow signals were then used to compute CL and R, by least mean square analysis.

Lung Volume Measurement Functional residual capacity (FRC) was measured with

the closed circuit helium dilution technique as described previously. Briefly, at end-expiration, the patient was connected to a closed circuit containing a known volume and concentration of helium. The circuit is adapted to maintain ventilator support during the mea- surement. The helium concentration decays for 90 sec- onds. The decay curve is biphasic; the initial rapid de- cline in helium concentration is due to equilibration with the infant’s FRC; second, slower decay is due to the steady leak of helium from around the endotracheal tube. The helium decay curve is then interpreted mathemati- cally to eliminate the loss of helium due to leak from that due to equilibration and arrive at a final helium concen- tration due only to the infant’s FRC. The computerized PEDS system was utilized for data collection, analysis, and storage.

Protocol The infants were randomized and placed on either CV

(n = 4) or HFJV (n = 5) for the first measurement in the study. The initial ventilator settings included an appropri- ate peak inspiratory pressure determined by the attending anesthesiologist (20-30 cmH,O), and 4 cmH,O positive end-expiratory pressure. The ventilator rate was set at 30 breathdmin with 0.5 second inspiratory time for CV (Ohmeda 78 10 ventilator, Madison, WI), and 420 breathdmin with 0.2 second inspiratory time for HFJV (Bunnell Life Pulse HFJ Ventilator Device, Bunnell Inc., Salt Lake City, UT). In addition, during HFJV, 10 sigh breathsimin were applied with the conventional ventila- tor. To maintain oxyhemoglobin saturation greater than 75 percent FiO2 was adjusted between 25 and 50%. After the induction of anesthesia, placement in the thoracotomy position, and adequate exposure of the surgical field, vital signs were recorded, and arterial blood gases, pul- monary mechanics, and FRC were measured. The infant was then placed on the alternate mode of ventilation (CV or HFJV) maintaining the same inspiratory and expira- tory pressures and FiO2 as for the previous ventilatory mode. When switching from CV to HFJV, a drop in airway pressures occurs until the appropriate servo-con- trolled driving pressure is obtained. To avoid a pressure loss to the infant, the CV rate was slowly diminished to sigh level, maintaining near stability in mean airway pressure. After several minutes, the rate was lowered to 10 sigh-breathdmin. Equivalent exposure of the surgical

lntraoperative Jet Ventilation 157

Fig. 1. Individual changes in mean airway pressure when changing from conventional (CV) to high frequency jet ventila- tion (HFJV).

field was attempted. After 10 minutes of the new mode of ventilation, and prior to surgical alterations in vessel anatomy, vital signs were recorded, and arterial blood gases, pulmonary mechanics, and FRC were measured. Photographs of the lung were taken without lung retrac- tion at end-inspiration on both CV and HFJV. The infants remained on the second mode of ventilation for the re- mainder of the procedure.

Data Analysis Differences in vital signs, arterial blood gases, pulmo-

nary mechanics, and FRC under different ventilating con- ditions were evaluated with the paired Student's t-test. Significant differences were accepted at P < 0.05.

RESULTS

The infants tolerated the shunting procedure well and were discharged from the hospital after mean 7 -t- 3 SD days postoperatively. The infants also remained stable during the transfer between CV and HFJV, data collec- tion, and the determination of pulmonary mechanics and FRC. The entire testing procedure required 3 to 5 minutes on each ventilator, and did not interfere with the surgical procedure.

The changes in MAP, P'lco2, and Pao2, with changing from CV to HFJV are displayed in Figures 1, 2, and 3, respectively. All infants experienced a decrease in MAP and 6/ 10 had a decrease in Pacq on HFJV when compared with CV. Four of ten infants had an increase, and three of ten had no change in Pao2 with HFJV when compared with CV.

Mean values for heart rate, systolic and diastolic blood pressure, mean arterial pressure, arterial blood gases, lung mechanics, and FRC, on CV and HFJV are shown in Table 1. There were no significant differences in vital signs, pulmonary mechanics, FRC, or Pao2 between modes of ventilation. As also shown, mean values for PaCo2 and MAP were lower with HFJV.

50

45

-40 m

& 30

'E 35

8 25

20

15

10 ~

cv HFJV

Fig. 2. Individual changes in Pa,, when changing from conven- tional (CV) to high frequency jet &tilation (HFJV).

30 X

20 + cv HFJV

Fig. 3. Individual changes in Pa, when changing from conven- tional (CV) to high frequency jet dentilation (HFJV).

Figure 4 is a photograph of the left lung inflated to FRC during operation and at the end of a tidal inspiration; the lung fills the surgical field at end-inspiration. As depicted in Figure 5 , during HFJV lung inflation is simi- lar at FRC and at end-inspiration, showing the same displacement as the lung at FRC while on CV. During the procedures the surgeons commented universally that ac- cess to the surgical field and minimization of lung retrac- tion was facilitated by HFJV.

DISCUSSION

The intraoperative management of infants undergoing cardiac surgery is complicated by pulmonary dysfunction and difficult access to the surgical field. The pulmonary dysfunction itself may be secondary to preoperative pul- monary morbidity or changes in pulmonary function due to the surgical manipulation of the lung and great ves- sels.13 In addition, the placement of the infant in the thoracotomy position, and retraction of the lung for ac- cess to the field may further compromise lung function and cardiopulmonary stability. Access to the surgical field for a Blalock-Taussig shunt procedure is limited by the small size of the infant's chest cavity, and by the presence of the lung over the region of interest. 'J This

158 Greenspan et al.

TABLE 1 -Measurements of Cardio-pulmonary Variables in Nine Infants on Conventional Versus High Frequency Jet Ventilationa

Measurements cv HFJV P value

Heart rate (beatsimin) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Mean airway pressure (cmH,O) Arterial pH Arterial Pco, (mmHg) Arterial Po? (mmHg) Lung compliance (mL/cmH,O/kg) Lung resistance (cmH,O/L/s) Functional residual capacity (mL/kg)

159 t 5 81 2 5 43 2 3

10.9 ? 0.91 7.40 2 0.03

37 ? 3.2 45.6 t 4.2 0.55 2 0.06 109 t I3 23 t 2

161 2 5 80 2 4 43 2 3

8.5 * 0.73 7.46 2 0.03

28 2 2.1 55.3 * 5.7 0.56 * 0.05

9 7 i 1 1 22.5 ? 3

NS NS NS

<0.01 0.052

<0.05 0.053 NS NS NS

~

“Data given as mean 2 standard error of the mean. CV, conventional ventilation; HFJV, high frequency jet ventilation.

Fig. 4. lntraoperative photograph of lung inflated to functional residual capacity (A) and at end-inspiration on conventional ventilation (B).

Fig 5. lntraoperative photograph of lung inflated to functional residual capacity (A) and at end-inspiration on high frequency jet ventilation (B). Camera angle and surgical exposure is simi- tar to that of Figure 4.

may be further exacerbated by lung inflation due to posi- tive end-expiratory pressure and tidal breathing move- ments, particularly when high inflation pressures are required to compensate for the aforementioned abnor- malities in pulmonary function.

High frequency ventilation uitilizes high respiratory rates and low tidal volumes to facilitate gas e ~ c h a n g e . ~ One mode of this therapy, HFJV, delivers a pulse of gas at a rapid rate via a specially designed endotracheal tube with an additional jet cannula placed in the proximal

lntraoperative Jet Ventilation 159

airway. Gas exchange occurs, in part, by a process of augmented diffusion along the respiratory tree. I4,l5 Ade- quate lung volume is maintained by positive end-expira- tory pressure and sigh breaths produced by a conven- tional ventilator placed in tandem with the HFJV equipment. High frequency ventilation has been demon- strated to maintain or improve gas exchange at a lower mean airway pressure in various neonatal lung disease states. 3-7 3 14-’ In addition, HFJV has been utilized intra- operatively to maintain a relatively motionless surgical field while maintaining gas exchange during adult airway surgery. ’

In the present study, nine infants were ventilated with both conventional and HFJV strategies during a shunting procedure. Vital signs and P,* remained similar on each mode of ventilation, with a lower Paco2 and a lower mean airway pressure on HFJV. The lower mean pressures result from the inspiratory/expiratory time ratio of 1:6 on HFJV. For this study, Pace was not controlled, and lower levels were observed at the same peak inspiratory and expiratory pressures on HFJV compared with CV. Sev- eral of the infants had Pace values that were outside the typical range (35-45 mmkg) on the initial ventilator. These ventilator settings were established by the attend- ing anesthesiologist, and Paco2 values may, in part, be the result of attempts to increase pulmonary blood flow, alter Pao2, or achieve other clinical goals. The potential effects of a lower PaC9 include a decrease in pulmonary vascular resistance, which may be beneficial to this population.” Had this protocol been designed to achieve matching levels of arterial Pace, rather than ventilator pressures on the two modes of ventilation, most likely the mean air- way pressure would need to be decreased further on HFJV, thereby increasing the difference observed in the present study. Further decreases in mean airway pressure with HFJV would potentially decrease barotrauma, air- way deformation, and pulmonary morbidity. ‘‘,19

The infants maintained similar pulmonary mechanics and FRC on each mode of ventilation. Hence oxygen- ation and alveolar distention should be similar on each ventilator. The lung volume at FRC was slightly lower than that of healthy term infants based on our experience. This was probably due to the presence of some intraoper- ative pulmonary compromise, and lung manipulation. l3 In addition, lung compliance was lower, and lung resis- tance higher than in healthy term infants. These latter abnormalities may have been due to preoperative and intraoperative lung dysfunction related to the cardiac dis- ease and manipulation, or to the use of a paralyzing agent during surgery. ’O,’ ’

In comparison to CV, tidal excursion are much smaller on HFJV. This results in diminished lung movement and, therefore, in less disruption of the operative field. In addition, the degree of lung expansion at FRC noted during CV does not occur in HFJV, minimizing the need

for extensive lung retraction. The surgical team observed superior access and stability of the surgical field on HFJV.

Alterations in pulmonary function have been reported in neonates following Blalock-Taussig shunting. I3**O In addition to changes in pulmonary blood flow from the surgical intervention, possible areas of pulmonary com- promise during this procedure include atelectasis of lung units from prolonged placement in the thoracotomy posi- tion under general anesthesia, and from lung retraction and manipulation.’’ The benefits of HFJV in ventilating areas of inhomogenously aerated lung parenchyma, are to produce more uniform gas e x ~ h a n g e . ~ * * ” ~ ~ Minimaliza- tion of lung manipulation facilitated by HFJ ventilation may also diminish intraoperative and postoperative pul- monary abnormalities. This point is purely speculative, however, because the short duration of the protocol ne- gated useful comparisons of surgical outcomes with HFJV versus CV, such as diminished intraoperative com- plications, shortened surgical time, the incidence of early graft closure, requirements for postoperative ventilatory support, etc. Such outcome variables need to be assessed in subsequent randomized studies. In addition to the in- traoperative utility of HFJV in the population studied, a speculative extrapolation to other thoracic or abdominal procedures, and clinical scenarios, is possible. In this regard, a more stable surgical field may be obtained with HFJV during open cardiac procedures or delicate bowel surgery. In addition, critically ill infants in need of a surgical intervention, managed in the intensive care unit on high frequency ventilation, need not be changed to CV for the procedure, as cardiopulmonary stability can be maintained in the operation with high frequency ventila- tion.

In conclusion, the use of HFJV during closed-heart cardiac surgical procedures in infants results in similar pulmonary function with improved gas exchange, at a lower mean airway pressure than CV, at the same inflat- ing pressures. This suggests that HFJV is an effective mode of intraoperative support for this patient popula- tion. In addition, some potential intraoperative clinical advantages of this modality of mechanical ventilation, may prove beneficial for postoperative outcome.

REFERENCES

1 . llbawi MN. Current status of surgery for congenital heart dis- eases. In: Reyes HM, Vidyasagar D, eds. Clinics in Perinatology: Neonatal Surgery. Philadelphia: WB Saunders, 1989: 157-176.

2. Del Nido PJ, Williams WG, Coles JG, Trusler GA, Freedom RM. Closed heart surgery for congenital heart disease in infancy. In Hastreiter AR, ed. Clinics in Perinatology: Cardiovascular Dis- ease in the Neonate. Philadelphia: W B Saunders, 1988:681-697.

3. Gerstmann DR, deLemos RA, Clark KH. High-frequency ventila- tion: Issues of strategy. In: Brans YW, ed. Clinics in Perinatology: Newer Technologies and the Neonate. Philadelphia: WB Saun- ders, 1991:563-580.

160 Greenspan et al.

4. Frose AB, Bryan AC. High frequency ventilation. Am Rev Respir Dis. 1987; 135:1363-1374.

5. Keszler M, Donn S, Bucciarelli RL, Alverson D, Hart M, Lu- nyong V, Modanlou HD, Noguchi A, Pearlman SA, Puri A, Smith DW, Stavis R, Watkins MN, Hams TR. Multicenter con- trolled trial comparing HFJV and conventional ventilation in new- born infants with pulmonary interstitial emphysema. J . Pediatr. 1991; 119:85-93.

6. Carlo W, Beoglos A, Chatburn R, Walsh MC, Martin RJ. High frequency jet ventilation in neonatal pulmonary hypertension. Am J Dis Child. 1989; 143:233-238.

7. Smith DW, Frankel LR, Derish MT, Moody RR, Black LE, Chipps BE, Mathers LH. High-frequency jet ventilation in chil- dren with adult respiratory distress syndrome complicated by pul- monary baratrauma. Pediatr Pulmonol. 1933; 15:279-286.

8. Borg U. HFPPV: Use in airway laryngoscopy and surgery under general anesthesia. Anesth Analg. 1982; 59:59L603.

9. Bhutani VK, Sivieri EM, Abbasi S, Shaffer TH. Evaluation of neonatal pulmonary mechanics and energetics: A two factor least mean square analysis. Pediatr Pulmonol. 1958; 4: 15@158.

10. Koen PA, Moskowitz GD, Shaffer TH. Instrumentation for mea- suring functional residual capacity in small animals. J Appl Phys- iol. 1977; 43:755-758.

1 1 . Goldsmith LS, Greenspan JS, Rubenstein SD, Wolfson MR, Shaffer TH. Immediate improvement in lung volume after exoge- nous surfactant: Alveolar recruitment versus increased distention. J Pediatr. 1991; 119:424-428.

12. Schwartz JG, Fox WW, Shaffer TH. A method for measuring functional residual capacity in neonates with endotracheal tubes. IEEE Trans. Biomed Eng. 1978; BME25:304307.

13. Greenspan JS, Davis DA, Russo P, Wolfson MR, Shaffer TH.

Operative creation of left to right cardiac shunts: Pulmonary func- tion sequelae. Ann Thorac Surg. 1993; 55:927-932.

14. Bunnell JB. High-frequency ventilation of infants. In: Current Perinatology, Vol. 11. New York: Springer-Verlag, 1990: 172- 200.

15. Wetzel RC, Gioia FR. High frequency ventilation. Pediatr Clin North Am. 1987; 34:15-38.

16. Spitzer AR, Butler S, Fox WW. Ventilatory response to combined high frequency ventilation and conventional mechanical ventila- tion for the rescue treatment of severe neonatal lung disease. Pediatr Pulmonol. 1989; 7:24&250.

17. Morray JP, Lynn AM, Mansfield PB. Effect of pH and pC0, on pulmonary and systemic hemodynamics after surgery in children with congenital heart disease and pulmonary hypertension. J Pedi- atr. 1988; 113:474-479.

18. Penn RB, Wolfson MR, Shaffer TH. Effect of ventilation on mechanical properties and pressure-flow relationships of imma- ture airways. Pediatr Res. 1988; 23:5 19-524.

19. O’Brodovitch HM, Mellins RB. Bronchopulmonary dysplasia. Unresolved neonatal acute lung injury. Am Rev Respir Dis. 1985; 132:69&709.

20. DiCarlo JV, Raphaely RC, Steven JM, Norwood WI, Costarino AT. Pulmonary mechanics in infants after cardiac surgery. Crit Care Med. 1992; 20:2@27.

21. Turnbull AD, Carlon G , Howland WS, Beattie EJ. High-fre- quency jet ventilation in major airway or pulmonary disruption. Ann Thorac Surg. 198 1 ; 32:468474.

22. Derderian SS, Rajagopal KR, Abbrecht PH, Bennett LL, Doblar DD, Hunt KK. High frequency positive pressure jet ventilation in bilateral bronchopleural fistulae. Crit Care Med. 1983; 10:119- 121.