A comparison of the effects of high frequency-low tidal volume and low frequency-high tidal volume mechanical ventilation

108

July 1980 The Journal o f P E D I A T R I C S

A comparison of the effects of high frequency-low tidal volume and low frequency'high tidal volume mechanical ventilation Ten neonates with severe hmg disease were studied while mechanically ventilated with standard volume preset infant ventilators, using two different ventilator), patterns. Slow venti[atory rates and high tidal volumes were alternated with rapid rates and low tidal volumes," minute ventilation, FIoe PEEP, and 1.'E ratios were held constant. Peak inspiratory pressure, mean airway pressure, expiratory time, Pao.,, Pac,=; pH, and arterial blood pressure were measured and compared for each frequency-tidal volume combination. The best arterial oxygenation occurred at the combination of settings that produced the highest mean airway pressure and always during low frequency-high tidal volume ventilation (P < O.001). Changes in oxygenation appeared to be directly related to changes" in MAP. A second experiment examined two different ventilator systems" responses to changes in ventilatory rate. When the rate o f one ventilator (Bourns LS104 volume prese 0 was increased, M A P increased. When the rate o f the other ventilator (Bennett PR2 pressure preset) increased, MA P decreased. These observations suggest that there is a direct relationship between M A P and arterial oxygenation, and that the supposed advantages" of one ventilatory pattern over the other may be secondary to inadvertent changes in subtle pressure-time relationships within the respiratory cycle and incidental changes in MAP. These changes may vary from one ventilator to another.

Stephen J. Boros, M.D.,* and Kathy Campbell, R.R.T.,

St. P a u l a n d

Minneapolis, Minn.

RECENT OBSERVATIONS by Bland et al 1 suggest that during the mechanical ventilation of infants with hyaline membrane disease, a ventilatory pattern of rapid shallow breathing or high frequency-low tidal volume ventilation may be superior to the more generally accepted pattern of low ventilatory frequencies and relatively high tidal volumes. These findings appear to contradict earlier studies by Smith et a l l ' 3 which showed that oxygenation varied directly with peak inspiratory pressure or tidal volume, and inversely with ventilatory rate. This report examines From the Children's Hospital, and Department o f Pediatrics, University o f Minnesota. Aided by the Research and Education Fund of Children's Hospital Read in part before the Annual Meeting o f The Midwest Society for Pediatric Research, Cincinnati, October 31, 1979. *Reprint address: Children's Hospital, 345 North Smith Ave., St. Paul, MN 55102.

VoL 97, No. 1, pp. 108-112

and compares the effects of the two different ventilatory patterns. PATIENTS

AND METHODS

Ten neonates receiving controlled mechanical ventilation were studied. Eight had HMD, one had a pulmonary Abbreviations used hyaline membrane disease HMD: I:E ratio: inspiratory:expiratory ratio 1=1%: Fraction of oxygen in inspired air positive end expiratory pressure PEEP: effective tidal volmne TVe: MAP: mean airway pressure partial pressure of arterial oxygen Pa%: partial pressure of arterial carbon dioxide Pae%: hemorrhage, and one had respiratory failure secondary to unexplained hydrops fetalis. All were paralyzed with pancuronium bromide and.ventilated with volume preset infant ventilators (Bourns LS104-150). More conservative

0022-3476/80/070108 +05500.50/0 9 1980 The C. V. Mosby Co.

Volume 97 Number 1

Comparison of mechanical ventilations

Table. Summary of clinical data .Mean • 1SD

Birth weight (gm) Gestational age (wk) Age studied (hr) FIQ PEEP (cm H20) I:E ratio Low frequency-high TV TVe (ml/kg) Rate (/rain) High frequency low TV TVe (ml/kg) Rate (/rain)

2,010 • 767 33.5 + 3 44 _+ 26 0.71 • 0.2 5.5 • 1.0 1:2.9

Range

1,305-3,950 30-40 13-96 0.45-i.0 4-7 1:2-1:3.4

9.2 • 1.5 33 + 5

6.6-12 25-40

4.6 + 0.8 66 _+ 10

3.3-5.3 50-80

A b b r e v i a t i o n s used: FIO._, = F r a c t i o n o f oxygen in inspired air; P E E P = positive e n d e x p e r a t o r y pressure; I : E ratio inspiratory: expiratory ratio; T V = tidal v o l u m e : T V e = effective tidal v o l u m e .

16~I

! 09

r

50

~ o ~ , ,~

-I~o ~--I~o ~o

ol

--I ~o~= --J ,o ,= E =E

I ~oo i90 180 E

170

o~

mo

16

140 0 ~ 130 n~ 120 I]O

12

I0 o: "~ I High freq. Low TV

I

Low Freq. High TV

. I, Low Freq, High TV

I High Freq. Low TV

Vent ilotory Pottern

forms ofventilatory support had failed in all patients. The clinical data are summarized in the Table. Initially, the ventilators were adjusted to deliver an effective tidal volume of 8-10 ml/kg. The effective tidal volume or delivered tidal volume is the volume set minus the volume lost within the ventilator and patient circuit. Initial rates were 25 breaths/minute. I:E ratios were 1:2 to 1:3.4, and were set using inspiratory flow rate alone. Inspiratory time hold mechanisms were not used. Inspired oxygen concentration, rate, tidal volume," and PEEP were adjusted to provide optimum oxygenation and ventilation. Thereafter, the only changes in settings were tidal volume and rate. This initial combination of settings was called low fi'equency high tidal volume and was considered baseline. The high frequency-low tidal volume condition was achieved by decreasing the TVe by 50% and doubling the baseline rate, maintaining baseline minute ventilation. With each change, inspiratory flow was readjusted to maintain a constant I:E ratio. At the beginning of each experiment, a steady state was achieved at the low frequency-high tidal volume baseline. Rates and tidal volumes were then varied in the following sequence: (1) low frequency-high tidal volume baseline; (2) high frequency-low tidal volume; (3) low frequency-high ,tidal volume baseline; (4) high frequency low tidal volume. Following a ten-minute equilibration period at each combination of settings, Pa%, Pa~:o2, pH, arterial blood pressure, peak inspiratory pressure, and mean airway pressure were measured. Because patients were ventilated at different oxygen concentrations, individual changes in oxygenation were standardized by the expression Pa%/ FI%.4-" Proximal airway pressures were measured at the operating head of the ventilator using a standard pressure transducer attached to a proximal airway-ventilator mon-

Fig, 1. Changes in arterial blood gases, blood pressure, and MAP associated with changes in ventilator settings (Pao_,/ FI% + SEM).

200 I

Low Freq

"._ - ' -~,160N 0 E

gh Freq LOWTV

g_v 140

p < 0.00/

120I84 11

I 12

I-

I

I

I

13

14

15

16

Mean Airway Pressure

(cm H20)

Fig. 2. Mean changes in Pa,,./FI,,., associated with changes in MAP. itoring system (Pneumogard, Novametrix Medical Systems). This system calculates MAP by sampling proximal airway pressures every 10 msec from the beginning of one inspiration to the beginning of the next, adding the respective sample points, and dividing by the number of samples. BlOod pressures were measured through umbilical artery catheters attached to standard pressure transducers (Hewlett-Packard HP1280C ) and standard processor display modules (Hewlett-Packard HP8205B). A separate experiment was carried out to investigate the effective changes in venti!atory rate on I:E ratio and MAP. Two different ventilators were examined. They were (1) a pressure preset, fixed I:E ratio ventilator (Bennett PR2), similar to the machine used by Smith et al-'

1 10

Boros and Campbell

The Journal of Pediatrics July 1980

O O

nILl ~

1:2 1:3 t:4 i:5

12 =

10 8

a'r

~E

6

O'l'D tr

2

o

Volume Vent. ,4 Flow ~ Pressure Vent

|

o Volume Vent.Constant F l o w I

"-// 20

1

I

I

I

I

J

30

a,'O

50

60

70

80

Ventilatory Rate(breaths per minute) Fig. 3. Changes in I:E ratio and MAP associated with changes in ventilatory rate using different mechanical ventilators. in 1969; and (2) a volume preset9 ventilator (Bourns LS104-150) identical to the machine used in our first experiment and similar to tl~at used by Bland et al 1 in 1977. Both ventilators were attached to the previously described ventilator monitoring system and a lung simulator (Bourns Model P/4500). Ventilator rates were increased in 10 bpm'increments and MAP, I:E iatio, and expiratory times were measured and compared. The pressure preset ventilator was set to deliver a peak inspiratory pressure of 25 cm H~O. The volume preset ventilator was set arbitrarily at 30 ml and an I:E ratio of 1:4. Tidal voldmes remained constant throughout the experiment. Two experiments were performed with the volume preset ventilator. (1) The ventilatory rate was increased while maintaining a constant iiaspiratory flow. (2) As the rate was increased, inspiratory flow was readjusted to maintain a constant I:E ratio. Statistical analysis was performed using a two-tailed t test. The project was approved by the Children's Hospital Committee on Human Experimentation. Informed parental consent was obtained. RESULTS Changes in mean arterial blood gas values, mean blood Pressures, and mean M A P in response to ch~inges in ventilator settings are displayed in Fig. 1. A listing of all ventilator variables, arterial blood gas measurements, blood pressure measurements, and proximal airway pressure measurements is available on request. As ventilator rates were increased and tidal volumes decreased, there was a decrease in MAP (P < 0.001) and

a concomitant fall in arterial oxygenation (P < 0.01). Pae% values rose slightly (P < 0 . 0 5 ) and arterial blood pressures did not change. After return to the low frequency-high tidal volume baseline, both Pa%/FI% and MAP returned to near the original baseline values. Increasing the rate and decreasing effective tidal volume a second time again was accompanied by a marked decrease in both Pao2/FI % and MAP. Fig. 2 shows the changes in Pa%/FI% values in terms of MAP alone. In all patientsl pao2/Fio._, increased as MAP increased. In all patients, the best oxygenation occurred at the highest MAP and, in all, t h e highest MAP was Observed during low frequency-high tidal volume ventilation (P < 0.001). Fig. 3 illustrates the results of the comparative ventilator trial. When the ventilatory rate of the volume preset ventilator Was progressively increased and inspiratory flow held constant, I:E ratio and MAP also increased pi'ogressively. When the inspiratory fl0w was readjusted with each change in rate to maintain a constant I:E ratio, MAP did not change Until the most extreme rate (80 bpm) was reached. At this combination of settings, it was not possible to maintain the preset I:E ratio, which increased from 1:4 to 1:3.5. Consequently,. MAP increased slightly. Increasing the rate of the pressure preset ventilator had little effect on I:E ratio or MAP until a rate of 40 bpm was reached. At rates above 40 bpm, both I:E ratio and MAP progressively decreased. Rates above 60 bpm could not be achieved with this machine: The shortest expiratoi~y time observed was 0.3 second, with the volume preset ventilator when inspiratory flows

Volume 97 Number 1

were not readjusted. W h e n flows were readjusted to maintain a constant I:E ratio, the shor!est expiratory time observed with this ventilator was 0.6 second. Th e pressure preset ventilator, s shortest expiratory time was 0.7 second. DISCUSSION The first clinical trials of intermittent positive pressure ventilation in the treatment of infants with H M D attempted to mimic distressed infants' natural respiratory pattern of rapid shallow breathing. This regimen of rapi d rates and modest tidal volumes was adequate for the treatment of hypercapnia, but inadequate in preventing hypoxia. 7 In 1969, Smith et al = examined the effects of independent variations of ventilatory rate an d peak inspirat0ry pressures, and observed that arterial oxygenation varied inversely with rate and directly w!th peak inspiratory pressure or tidal volume. These, and similar observations by Reynolds,7 led to a widespread acceptance of the ventilatory regimen of slow rates and relatively high tidal volumes. In 1977, Bland et al 1 described the successful mechanical ventilation of 23 infants with HMD, using low tidal volumes and very rapid rates. Twenty-0ne infants (91%) survived. Why was this ventflatory pattern unsuccessful in 1969 and successful in 19777 During their early investigations, both Smith and Reynolds used pressure preset ventilators with relatively fixed I:E ratios (Bennett PR2). As we observed, when the rate of this machine is increased, I:E ratios and MAP remain faMy constant until rapid rates are reached. Then, both MAP and I:E ratio decrease progressively. The majority of the patients in the Bland et al study were ventilated with volume preset infant ventilators with I:E ratios that varied With inspiratory flow rates, w e also observed that if inspiratory flow rates are not constantly readjusted, as ventilatory rates increase, I:E ratios and MAP increase progressively. High frequency ventilation, in the study of Smith et al in 1969, was associated with deteriorating oxygenation, perhaps because the mean airway pressure decreased, whereas in the Biand et al study, the fast ventilatory frequencies may well have been associated with increased mean airway pressure and, therefore, increased oxygenation. Mean airway pressure is a composite measure of peak inspiratory pressure, inspiratory time or I:E ratio, and PEEP. If these variables are kept constant and only the ventilatory frequency is altered, only total ventilat0ry time changes. Since neither pressures nor pressure-time relationships change, the mean airway pressure is not affected. Aside from the changes observed in proximal airway

Comparison o f mechanical ventilations

111

pressures, the combination of rapid rates, elevated I:E ratios, and shortened expiratory times can produce inadvertent pressure changes within the distal airways. If expiratory time becomes shorter than a given system's critical emptying time or time constant, that system will not empty completely. Gas will be trapped and pressures in the distal airways will progressavely increase? 9 In our study, this appeared most likely to occur with the volume preset ventilator operating at very rapid rates and elevated I:E ratios. Previous studies from this institution demonstrated the value of proximal MAP measurements as a composite measure of all pressures transmitted to the proximal airways by a mechanical ventilator? ~ 1~ Those studies. and earlier work by H e r m a n and Reynolds. 1~ showed a direcl relationship between MAP and arterial oxygenation. The present study again shows a direct relationship between MAP and arterial oxygenation. Our observations also suggest that the supposed advantages of one frequency-tidal volume ventilatory pattern over the other may be secondary to inadvertent alterations in subtle pressuretime relationships within the respiratory cycle and ramdental changes in MAP. REFERENCES

1

2.

3.

4.

5.

6. 7.

8.

9. 10.

Bland RD. Kim MH. Light MJ. and Woodson JL: Highfrequency mechanical ventilation of low birth weight infants with respiratory failure from hyaline membrane disease. Pediatr Res 11:531. 1977 (abstrl. Smith PC. Daffy WJR. Fletcher G. Meyer HBP. and Taylor G: Mechanical ventilation of newborn infants. I. The effect of rateand pressure on arterial oyygenation of infants with respiratory distress syndrome. Pediatr Res 3:244. 1969. Smith PC. Schach E. and Daily WJR: Mechanical ventilation0f newborn infants. II. Effects of independent variation of rate and pressure on arterial oxygenation of infants with respiratory distress syndrome. Anesthesiology 37:498. 1972. Gilbert R. and Keighley GR: The arterial/alveolar oxygen tension ratio. An index of gas exchange applicable to varying inspired oxygen concentrations. Am Re9 Resp Dis 109:142. 1974. Kirby RR. Downs JB. Civetta JM. Modell JH. Dannemiller FJ. Klein EF and Hodges M: High level positive end expiratory pressure I PEEP) in acute respiratory insufficiency, Chest 67:156. 1975. Horovitz JH. Carrico CJ. and Shires T: Pulmonary response to major injury, Arch Surg 108:349. 1974. Reynolds EOR: Pressure wave form and ventilator settings for mechanical ventilation in severe hyaline membrane disease. Intern Anesth Clin 12:259. 1974. Weigl J: The infant lung: A case against high respiratory rates in controlled neonatal ventilation. Resp Therapy 3:57. 1973 Donahue LA. Thibeault DW: Alveolar gas trapping and ventilator therapy in infants, Resp Therapy 9:47, 1979. Boros SJ, Matalon SV, Ewald R, Leonard AS, and Hunt

1 12

Boros and Campbell

CE: The effect of independent variations in inspiratoryexpiratory ratio and end expiratory pressure during mechanical ventilation in hyaline membrane disease: The significance of mean airway pressure, J PEDIAI"R 91:794, 1977. 11. Boros SJ: Variations in inspiratory-expiratory ratio and

The Journal of Pediatrics July 1980

airway pressure wave form during mechanical ventilation: The significance of mean airway pressure, J PEmATR 94:114, 1979. 12. Herman S, and Reynolds EOR: Methods for improving oxygenation in infants mechanically ventilated for severe hyaline membrane disease, Arch Dis Child 48:612, 1973.

Brief clinical and laboratory observations Parathyroid hormone in hypocalcemic and normocalcemic infants of diabetic mothers Akihiko Noguchi, M.D., Mustafa Eren, B.A., and Reginald C. Tsang, M.B.B.S., Cincinnati, Ohio

TH E R E I S a high incidence of hypocalcemia in infants of diabetic mothers. 1 a Functional hypoparathyroidism is thought to be a cause of hypocalcemia in these infants, :' but parathyroid function has not been examined in hypocalcemic as compared with normocalcemic IDM. In the present study, we examined postnatal serum parathyroid hormone changes in persistently hypocalcemic IDMs compared with normocalcemic IDMs, in order to determine whether parathyroid function is different in these two groups of IDM. PATIENTS

hour-old neonates were drawn for determinations of serum total calcium and ionized calcium, magnesium, phosphorus, ~ and parathyroid hormone. The PTH radioimmunoassay mainly detects N-terminal portions of the polypeptide and detects 89% of normal adults? Student t test, rank t test, paired t test, multiple regression, and analyses of covariance were used for statistical analyses. Abbreviations used IDM: infantsof diabetic mothers PTH: parathyroid hormone iCa: ionized calcium

AND METHODS

Forty-six IDMs and their mothers admitted to the University of Cincinnati Hospital, General Division, were enrolled in the study after informed consent was obtained. M~thods of newborn care, determination of gestational age, and methods of blood sampling in this hospital were previously describedY Maternal blood at delivery, umbilical venous blood, and blood from 24-, 48-, and 72-

From the Department of Pediatrics, Newborn Division, the Department of Obstetrics and Gynecology, the Crosley Memorial Nursery, Cincinnati General Hospital, University of Cincinnati College of Medicine; and the Children's Hospital Research Foundation. Supported in part bfi NIH NICHD HD 11725, Children's' Hospital Research Foundation, NIH NICHD CRC Grant RRO0123, GCRC Grant RR68, and H E W Maternal and Child Health Training Grant Project No. 174.

RESULT In 46 IDMs, 27 infants had serum Ca concentrations <7.5 m g / d l at least once during the period of study (21 infants <7.0 m g / d l ). Nine patients had serum Ca concentrations <7.5 mg/dl at 24, 48, and 72 hours of age, hereafter called the "hypocalcemic IDM group." In ten patients the serum Ca levels were always >7.5 m g / d l at 24, 48, and 72 hours of age, hereafter called the "normocalcemic IDM group." The hypocalcemic a n d normocalcemic groups of IDM are the main focus of the present study. Clinical information regarding the two groups of IDM are presented in the Table. Hypocalcemic IDM were more premature and had severer degrees of maternal diabetes (White classification). One-minute Apgar scores in the two groups were similar. Serum total calcium concentra0022-3476/80/070112+03500.30/0 9 1980 The C. V. Mosby Co.