NEONATOLOGY ON THE WEB

Historical Review and Recent Advances
in Neonatal and Perinatal Medicine

Edited by George F. Smith, MD and Dharmapuri Vidyasagar, MD
Published by Mead Johnson Nutritional Division, 1980
Not Copyrighted By Publisher

Chapter 18

Historical Perspectives and Recent
Advances of Neonatal Monitoring

Dharmapuri Vidyasagar, M. D.

Monitoring of the newborn is an old biological phenomenon. All species of the animal kingdom monitor their offspring from birth, using the sense organs for smelling, tasting, feeling, seeing and hearing (Table 1). In medicine, for centuries, touching and feeling were the methods used. Later, methodical observations by smell and taste were included in clinical monitoring. Direct auscultation of the chest, as mentioned by Hippocrates, was not used as a major mode of examination. The first instrument introduced into clinical monitoring was probably the thermometer. The stethoscope, another instrument for monitoring, appeared only about 165 years ago. Modern methods of electronic monitoring are less than 100 years old. It is difficult to ascertain exactly when the first electronic monitor was used in a clinical setting and even more difficult to find such information regarding neonatal monitoring. There seems to be a lag of ten years between the introduction of an electronic monitor into a physiologic laboratory and the clinical use of it. Most of the current monitoring systems were first used in physiology laboratories and later in cardiac catheterization laboratories. Finally, they were introduced into day-to-day clinical usage.

In 1968, an article in Hospital Progress^[1] captioned "Medical First After Thermometer and Stethoscope" reported that the newborn unit at Cardinal Coleman Hospital, St. Louis, Missouri, had established total neonatal electronic monitoring systems for clinical purposes. The report stated "space age technology and the determination of a St. Louis pediatrician to lower infant mortality has created the first neonatal monitor in the world." Julius Comroe,^[2] recounting the historical perspectives of the progress made in conquering hyaline membrane disease stated that "Day in Kansas had organized the first Coronary Care Unit in 1962-63, and Mildred Stahlman at Vanderbilt had developed the first genuine Neonatal Intensive Care Unit about 1963-64. William Tooley started his about 1965-66." Medical historians may question the chronological order, nevertheless, the report suggests that until 1968 electronic monitoring had not yet become routine in neonatal country, USA.

The progress in neonatal monitoring has been relatively slow and continues to lag behind developments in the adult fields. There are several reasons for this: First, neonatal medicine is a younger scientific field. Second, neonatal monitoring requires miniaturization of existing technology and special designs because of its unique physiological differences. Finally, the manufacturers have been reluctant to invest in products having a slow sales volume. However, it is obvious that during the last decades neonatal monitoring has expanded exponentially. It is heartening to see that the explosion of space age technology has led to the development of minitransducers and microprocessors which facilitated the development of neonatal monitoring devices. The progress has been rapid and today's neonatal ICU is getting to look more and more like the inside of a spaceship.

Monitoring of the neonate can be classified into three major categories: (a) clinical (b) biochemical and (c) electronic. Clinical monitoring is mainly based on clinical findings of inspection, palpation and auscultation. Biochemical monitoring is mainly dependent on blood withdrawal and biochemical analysis. Noninvasive biochemical monitoring is being used more frequently. Electronic monitoring used primarily for monitoring physiological variables can either be invasive or noninvasive. This chapter will report on the historical development of the instruments used in neonatal monitoring.

THERMOMETER

Assessing body temperature is a basic clinical procedure in medical practice. In the early days, it was thought that the patient's pulse was more important than body temperature in assessing wellbeing. Today we know better and in the neonate, in particular, temperature maintenance is of prime importance.

The first thermometer was made by Galileo in 1592, but the clinical use of thermometry was introduced by Herman Boerhaave in 1668-1788. Today's clinical thermometer was designed by Allbutt in 1870. The next important gain came about in 1961^[3] when the servo controlling mechanism was introduced to maintain temperature in the incubators using a direct skin temperature measurement. Later, a simpler technique of measuring temperature using a temperature tape was introduced, but it did not gain popularity. More recently, the introduction of electronic temperature measurement on a regular basis has facilitated temperature monitoring in many ways. These electronic thermometers are fast-reading and easy to handle and there is no problem of breakage and spilling of mercury. The concept of continuous monitoring of temperature, i.e. differences between skin and toe temperature, has been utilized to monitor the neonatal state of peripheral perfusion.

STETHOSCOPE

Until the stethoscope was introduced by Laennec, heart and breath sounds were heard by direct listening against the chest. Laennec's invention of the stethoscope is attributed to his observation of two children playing with wooden boards transmitting sounds from one side to the other. Figure 1 shows Laennec listening directly to the chest of a patient (1816). Feer's Textbook of Pediatrics,^[4] published in 1922 shows the direct method of auscultation being used on an infant (Figure 2). It is interesting that it took almost 100 years to replace direct auscultation with the stethoscope. It is clear that Hess auscultated the lungs of premature infants,^[5] but there is no indication of the wide use of stethoscopes in neonates. Silverman^[6] stated that "auscultation of the chest is difficult with an ordinary stethoscope." He recommended the use of a stethoscope electronically amplified (Figure 3), which was designed by Dr. Day.

BLOOD PRESSURE

Stephen Hales was the first to measure blood pressure using invasive techniques, in 1733. In 1876, Von Bosch invented an instrument which could measure blood pressure in humans noninvasively, thus originating sphygmomanometer. The instrument was refined by Scipione Riva Bocci, in 1896. Blood pressure was measured feeling the pulse while inflating. In 1805, Kortokoff used the stethoscope to monitor the pulse. Application of ultrasound techniques to measure infants blood pressure became available in the late 1960's.

Blood pressure in the neonate has been measured by many different techniques. Vieroidt^[7] is noted to have made the first blood pressure measurement in the newborn in 1894. Earlier instruments included Gartner's Tonometer and Riva Rocci's sphygmomanometer. The most extensive study in newborns was that of Ballard^[8] who studied blood pressure from birth, at one hour intervals for 12 hours, then each day for ten days. In addition, he recorded pulse rate, temperature, and body weight. Using oscillatory methods, Ballard is noted to have successfully recorded infants as small as 920 grams. Rucker and Connell^[9] and Reis and Chaloupka^[10] separately published blood pressure measurements in a large number of term infants, from birth to 10 days of age. Following these studies, there is a gap until Londe^[11] reported the same observations in preterm infants.

Meanwhile, the technique of blood pressure measurement had improved further. In 1952, recognizing the difficulties involved in auscultatory methods of measuring, Goldring and Wohltmann^[12] reported a simple flush method of blood pressure determination. Forfar,^[13] four years later, published extensive data on 413 measurements of 143 infants. It was not until 1969 that the importance of intra-arterial measurement of blood pressure in the neonate became apparent. Phibbs^[14] noted that, whereas in the adult and older children with circulatory or respiratory collapse, blood pressure was closely monitored, such was not the case in neonates. Through his extensive work, routine intra-arterial as well as noninvasive blood pressure has become a standard procedure in the NICU.

MONITORING OF RESPIRATIONS

Breathing activity is known to be the sign of life from the beginning of mankind. However, there has been little said of respiration or its pattern even in Laennec's treatise on diseases of the chest (1821).^[15] Much less was studied in the neonate until the middle of the 20th century. Although Miller^[16] in 1952 studied respiratory patterns using electronic equipment, no mention was made of quantitating the frequency of respiration. In 1955, Miller^[17] expressed his concern regarding the lack of methods for estimating the severity of respiratory insufficiency and he noted that there were no objective methods of quantitating respiratory insufficiency. He, therefore, went on to study respiratory rate patterns in the newborn infant. According to him "routine observations were scheduled every 15 minutes for the first hours, and every two hours for the next 48 hours." Thus he established, for the first time, clinical monitoring of respiration in the newborn.

The fact that premature infants have recurrent attacks of apnea was well established by Miller. The association of recurrent attacks of apnea and high mortality was recorded by Reid and Tunstall.^[18] The instrument for continuous monitoring of respiratory rate was only introduced in 1961.^[3] Finally, the era of the neonatal intensive care unit had arrived. Through special grants provided for the study of neonatal physiology, investigators had established elaborate laboratories equipped with the first generation of electronic equipment. Sometimes there were debates for continuous respiratory rate monitoring versus heart rate monitoring. Because of the high cost of equipment not all infants had both heart rate and respiratory rate monitoring. In Dunham's Premature Infant,^[6] first revised and expanded by Dr. Silverman in 1961, no mention was made of continuous heart rate or respiratory rate monitoring!

In 1967, Wick and Schmitt^[19] developed a technique for recognizing apneic attacks. Prior to their report, only clinical recognition was possible. A decade later, respiratory monitoring had become more complex with the introduction of continuous 24 hour monitoring for recognition of apnea, its frequency and pattern.^[20] The respiratory monitoring devices are mainly based on the principle of the impedance technique, measuring the changes in electrical resistance during respirations. Other simpler, noninvasive techniques based on vibrations were also introduced.

Simultaneous progress was made in heart rate monitoring. Electronic heart rate monitoring seems to have started earlier, but it did not enter into routine clinical practice until the 1960's. Smith,^[21] an obstetrician, studied the heart rate in mothers, in 1922, using EKG. As a matter of curiosity, he extended his observations to the newborn infant immediately following birth. He stated that EKG, besides giving an electrical tracing of the heart, could also be used to measure the heart rate. Prior to this study, there were few reports of EKG of newborns and infants; none were done immediately after birth, nor did they indicate that heart rate monitoring could be done using EKG.

The well written book, The Physical and Mental Growth of Prematurely Born Children, edited by Hess, Mohr and Bartelme^[5] in 1934, indicates that heart rate monitoring using an EKG recording had not yet arrived. Londe,^[11] in his book, describes his experience with EKG on 25 preterm infants who had average weights of 2132 grams (1485-2580 gram range). The youngest infant was two days old; twenty infants were less than 24 days. He stated that "as far as 1 know no statement concerning the heart rate in premature infants occurs in the literature." He further writes "as a rule, we were able to count pulse rate of cardiac auscultation." Was he referring to the earliest use of the stethoscope in the neonate? Many workers, until that time and even later, described pulse rate, not heart rate, and pulse rate by palpation of peripheral vessels.

Again, in the textbook, The Premature Infant, published in 1941, Hess^[22] mentions little on the use of a stethoscope for auscultation of heart rate or respirations. In 1961, Silverman^[6] reported no continuous recording but the stethoscope is mentioned for auscultation of heart rate and respiration. About this time, electronic monitoring had slowly progressed into the cardiac catheterization laboratories, but it was not yet put to clinical use.

O₂ ADMINISTRATION AND MONITORING OF INSPIRED O₂ CONCENTRATION

Subsequent to controlling the thermal environment, administration of oxygen was the major step to altering the neonatal environment. Administration of oxygen as a method of combating respiratory problems following birth was described as early as 1900 in the pediatric literature. The methods of administration and methods of monitoring were vague. The earliest method of quantitation of O₂ administration was to count the number of bubbles when O₂ was delivered through water. Until the 1960's, O₂ was merely administered in terms of liters/minute rather than by concentration. In 1961, Dr. Silverman,^[3] in his textbook, for the first time mentioned that "oxygen concentration must be determined by means of oxygen analyses, as often as necessary, to keep it properly stabilized, but at least every four hours . . ." In earlier days, the Beckman's paramagnetic O₂ analyzer was used for intermittent FiO₂ measurements. Since then numerous O₂ sensors have been developed for continuous measurement of inspired O₂ concentrations. Now it is mandatory that every infant receiving supplemental O₂ be monitored accurately and the procedure recorded properly.^[23]

MODERN MONITORING

In recent years, the emphasis has been placed on expanding neonatal monitoring to include physiological parameters, and on making the monitoring noninvasive and simpler to deal with. The most outstanding examples are those of noninvasive blood gas monitoring, monitoring of intracranial pressure and cerebral blood flow, and noninvasive biochemical monitoring.

TRANSCUTANEOUS BLOOD GAS MONITORING

Transcutaneous (tcpO₂) monitoring is indeed a major technological advance of the decade of the 1970's. The application of a miniaturized and heated Clark electrode to measure transcutaneous O₂ was simultaneously pub lished in 1972 by Huch and Huch et al.,^[24] and Eberhard et al.^[25] This historic event opened the way for a practical method of monitoring blood gas changes using noninvasive techniques. Almost immediately, these techniques were introduced into neonatal clinical use. This is one area of medical technology in which the neonatologists had taken the lead over others. The fact that the transcutaneous pO₂ and pCO₂ monitoring truly tracks the physiological changes in the neonate has been proved by several workers. It is accepted that under stable conditions, transcutaneous pO₂ and pCO₂ values correlate well with arterial values. In unsteady states, these values cannot be regarded as substitutes for blood gas monitoring. However, extensive laboratory and clinical studies have demonstrated that transcutaneous blood gas data provide extremely valuable information during both steady and unsteady states. Thus, transcutaneous blood gas monitoring has come to stay.

Another area of progress has been in tissue pH monitoring.^[26] Although it cannot be totally considered as a noninvasive technique, this simple semiinvasive method of implanting a 1 mm long, 2 mm wide, glass tip into subcutaneous tissue has proved to be extremely useful in continously following the neonatal acid-base status. A combination of the above three electrodes will eventually make the noninvasive blood gas monitoring a simple procedure.

INTRAVASCULAR PO₂ MONITORING

In order to minimize the number of blood withdrawals and also to obtain continuous information regarding arterial p02, intravascular pO₂ electrodes were developed and used clinically.^[27] These electrodes are found to be extremely useful in following the pO₂ changes and trends. In the future, it is possible to have a multi-sensor intravascular instrument to measure not only blood gases, but also various biochemical parameters.

TRANSCUTANEOUS BILIRUBIN MEASUREMENTS

Although neonatal biochemical monitoring has been covered in a different chapter of this book, discussion of bilirubin measurement is included here because of the new technology involved in noninvasive instantaneous bilirubin measurement. The transcutaneous bilirubinometer is a reflectometer which permits the noninvasive monitoring of bilirubin levels. Yamanouchi et al.^[28] were the first ones to report their experience, in 1980. Subsequently, Heygi et al.^[29] confirmed the usefulness in evaluating the bilirubin levels in white and black infants in the USA. This technique seems to have great promise for clinical use.

VENTILATOR MONITORING

Hitherto, monitoring consisted of only counting the pressure generated, the end expiratory pressure and the inspired O₂ concentration. With increasing sophistication in our approach to neonatal ventilation, it is obvious that ventilatory monitoring has to be further improved. Again, noninvasive techniques are necessary in order to be applicable in the neonate. Such systems are in the process of development but are not yet ready for clinical use.

Mass spectrometry has been shown to be of some use in measuring expired gases continously, thus providing data regarding expired O₂ and CO₂. Used in conjunction with blood gas data, this information will provide a great deal of physiologic information.

Similarly, pulmonary compliance, resistance, minute ventilation and functional residual capacity can be monitored using noninvasive technology. These techniques, when available, will be of immense value.

NONINVASIVE ICP MONITORING

Attempts to determine the intracranial pressure (ICP) by direct measurement via a trephined opening were described as early as 1866.^[30,31] These studies were restricted to animals. Corning^[31] was the first one to puncture the subarachnoid space of a living person in 1865. Since the introduction of simple methods of lumbar puncture using a needle, by Quincke^[32,33] in 1891, single or repeated readings of the spinal fluid pressure have been monitored in neurosurgical patients by placement of an intraventricular catheter,^[33] a subdural screw,^[34] subarachnoid switch,^[35] and intradural pressure transducer.^[36] In pediatric patients, lumbar puncture is used for obtaining cerebrospinal fluids for diagnostic studies and simultaneous monometric measurements of the pressures. In recent years, invasive techniques for measuring intracranial pressure have been utilized in the management of Reyes syndrome.^[37] In the newborn infant, insertion of the needle into the saggital sinus has been used to measure ICP by Vert et al.,^[38] however, the general practice of assessing the measurement of intracranial pressure in newborn infants is by palpation of the anterior fontanelle. Invasive techniques such as lumbar or ventricular punctures can not be applied routinely. On the other hand, palpation alone is not adequate to document subtle changes that occur in the activity of the neonate. Various noninvasive techniques to measure ICP in the newborn and infants have been reported. Principal among them are modifications of the Shiotz tonometer,^[39] oscillographic technique,^[40] applanation method,^[41] and a modified stethoscope diaphragm with pulse pick-up.^[42] None of these techniques was used routinely to monitor ICP in sick neonates. In fact, there is no awareness of the importance of monitoring ICP in the newborn and infants on a routine basis. Since the neonate has an advantage of an open fontanelle through which the pressure changes within the cranium are transmitted, we use the anterior fontanelle for measurement of pressure using the transducer developed by Numoto and associates.^[35] They developed an implantable switch to record ICP. The device made possible a balancing of the ICP acting against the implanted gold electrodes by an external force of air. A switch mechanism indicated the balance point and thus the ICP. Later, Epstein et al.^[43] modified this system and devised a pressure activated electro-optically controlled servo-mechanism to record the ICP by the intracranial placement of the electrode. We utilized the same fiberoptic sensor, extracranially, placed over the anterior fontanelle, to obtain anterior fontanelle pressure (AFP) in different clinical states.^[44]

AFP monitoring has been investigated by other workers as well and has shown promise as an important adjunct in patient management.

COMPUTERIZED TOMOGRAPHY AND ULTRASOUND SCANNING

Computerized tomography and ultrasonography were introduced into neonatal application only recently, i.e., the late 1970's and early 1980's.^[45] Using the anterior fontanelle as a window, these techniques have been extensively used for neonatal neurological monitoring. Blood flow velocity in the anterior cerebral arteries have been measured using the Doppler techniques.^[46] Ultrasound has been used to image the intracranial contents.^[47] Thus, in the near future, it should be possible to obtain elaborate physiological information regarding central nervous system states in the neonate without invasive techniques.

IATROGENIC PROBLEMS OF MONITORING

Every new method of monitoring introduced into neonatal nurseries has carried certain risks or complications.^[48] These complications have been well described and discussed previously by many investigators. The major complications of monitoring are related to the invasive nature of some of the procedures. Clinical monitoring itself, because of its limitations, is filled with pitfalls, e.g., O₂ administration on the basis of skin color or relying on respiratory rate to assess pCO₂ whereas invasive techniques of placing arterial catheters or performing arterial punctures have the inherent problems of introducing infectious agents, causing rupture of vessels, or leading to thromboembolic complications. On a different note, electronic devices, unless properly grounded or checked, may have electrical leakages severe enough to cause cardiac arrythmias or arrest.

Thus, noninvasive techniques have an immense edge over invasive procedures. These are particularly significant in the neonate because of the size of the infant and the infant's volume. However, one has to be content with the limitations of noninvasive monitoring in obtaining true physiological information.

FUTURE OF NEONATAL MONITORING

It can be projected that both invasive and noninvasive monitoring will rapidly progress. Emphasis will be placed on noninvasive systematic monitoring of cardiac, pulmonary and central nervous system functions. Invasive and noninvasive electrodes with multiple sensors will replace many biochemical tests. Computerization of physiological, biochemical and patient management data will become a reality. Attempts will be made to develop closed loop systems within the monitors to effectively treat the infants at the instant of physiological or biochemical derangements.

Dr. Nicholas Nelson once asked, "Who shall monitor the monitor?" This is a highly pertinent question in present day technology. Since there is a multitude of electronic equipment surrounding the critically ill, tiny, premature infant, it is virtually impossible for a nurse or a physician to watch the monitors or for that matter follow the monitors closely. In order to instill some reasoning and logic, it is essential that we develop computer assisted systems to monitor the monitors. This is not a far cry from reality. Several methods of innovative desk top and commercial computer systems are being introduced into neonatal management systems. These will soon provide some form of order and logic for the clinicians to monitor the monitors.

THE EXTENT AND COST OF MONITORING

It is estimated that there are 448 neonatal intensive care units in the USA as of 1979 (Table 2). Among these, the tertiary care center beds number 3700 and the level II center beds number 6340. The total number of ICU beds amount to 10,040.^[49] Thus, the extent of neonatal monitoring required can be estimated. In addition, monitoring devices are required in the delivery rooms, operating rooms, cardiac catheterization rooms and other intensive care areas.

The development of modern neonatal intensive care units across the nation has raised several questions concerning the cost and the effectiveness of the intensive care system. Sinclair et al.,^[50] in a recent analysis of the problem, stated that "neonatal intensive care programs require further evaluation with rigorous scientific methods." Although there are several pieces of information available regarding the effectiveness of a given procedure, there are no data to validate that intensive care and intensive care monitoring have unequivocally improved the neonatal outcome. However, Kitchen,^[51] in a 1971 study demonstrated that introduction of intensive care procedures into his unit led to a decrease in neonatal mortality; and Kleinman et a1.^[52] and Lee et a1.^[53] state that the most likely and plausible causes for the decline in neonatal mortality are the improvement in medical care and medical technology.

The entrance of space age technology, on the other hand, has caused two major concerns. One is that of cost; another is that of the rapidity with which electronic monitoring is changing and improving. The dreadful aspect is that clinical health personnel and hospital administrators are left with difficult but necessary decisions regarding the efficacy, efficiency and exigency of the monitoring systems to be installed in their respective units.

When prudently planned, and when both the consumer and the vendor are working on faith and personal commitment, the assimilation of space age technology will become simple. Fortunately, many vendors strive to achieve this through their personal involvement and commitment.

The cost of neonatal monitoring itself is unknown. Similar to the observations of Sinclair et al.^[50] in regard to perinatal care, a great amount of scientific work has to be accomplished to establish the efficacy and cost of electronic neonatal monitoring.

Making certain assumptions, we attempted to estimate the cost of electronic monitoring prorated per day, per bed. These values are shown in Tables 3a and 3b. It is assumed that a maximum care NICU bed should have monitoring units to measure (1) heart rate, (2) respiratory rate, (3) temperature, (4) inspired 02 concentration, (5) blood pressure, (6) transcutaneous 02, (7) transcutaneous C02, (8) tissue pH, (9) intracranial pressure, and (10) pulmonary arterial pressure. A conservative estimate of such monitoring would put the cost of the equipment at $36,000. Calculating at a 3 year depreciation, the cost per bed would be $12,000 per year. Based on the fact that each bed is occupied, on the average, for 15 days, it can be calculated that each bed would provide care for 24 infants per year. Thus, per diem, per bed cost would work out to be about $33.

Using similar calculations and adjusting for fewer monitors and higher use of each bed, the per diem cost of monitoring would be $10 per day. These costs are far less than one would think, considering the enormous benefits of intensive care on the outcome of the infant. Elsewhere in this book the definitive impact of neonatal intensive care has been clearly demonstrated. Thus, there is ample evidence that modern neonatal monitoring has remarkably improved our ability to track the sick neonate. This ability has improved the effectiveness of various therapeutic measures that have been adopted in the NICU.

In summary, neonatal monitoring has come a long way in the past two decades. We are just beginning to see some of the most innovative yet unexplored technological possibilities. The next two decades will be an exciting era to watch.

	Fig. l. Drawing showing Dr. Laennec examining his patient by direct auscultation.
	Fig. 2. Sketch of a figure, taken from Feer's T xtbook of Pediatrics (1922), showing the direct method of auscultation being used on an infant.
	Fig. 3. Drawing of stethoscope electronically amplified for neonatal use by Dr. Day.

REFERENCES

1. A medical first as important as the thermometer and stethoscope: The newborn monitor. Hospital Progress June 1968, p. 26.

2. Comroe J. H., Jr.: Retrospectroscope. Menlo Park Calif: Von Gher Press, 1977.

3. Monitor International. Narco Health Companies, vol. 6, 1972.

4. Feer E.: Textbook of Pediatrics. Translated by Sedgwick P., Sherer B. Philadelphia: Lippincott Company, 1972, p. 79.

5. Hess J. H., Mohr G. J., Bartelme P. F. (eds.): The Physical and Mental Growth of Prematurely Born Children. Chicago: The University of Chicago Press, 1934.

6. Silverman W. A.: Dunham's Premature Infants, 3rd ed. New York: Paul B. Hoeber, Inc., 1961.

7. Vieroidt B.: Daten and Tabellen. Ed. 2, Jena, 1893.

8. Balard P.: Quoted in the article by Londe (Ref. 11).

9. Rucker M. P., Connell J. W.: Blood pressure in the newborn. Am. J. Dis. Child. 27:6, 1924.

10. Reis R. A., Chaloupka A. J.: Blood pressure in the newborn following normal and pathological labor. Surg. Gynecol. Obstet. 37:206, 1923.

11. Londe S.: Studies of blood pressure, electrocardiograms, pulse rate, and roentenograms of the heart in premature infants. In Hess J. H., Mohr G. J., Bartelme P. F. (eds.): The Physical and Mental Growth in Prematurely Born Children. Chicago: University of Chicago Press, 1934, p. 277.

12. Goldring D., Wohltmann H.: Flush method for blood pressure determination in newborn infants. J. Pediatr. 40:285, 1952.

13. Forfar J. 0., Kibel M. A.: Blood pressure in the newborn estimated by the flush method. Arch. Dis. Child. 31:126, 1956.

14. Phibbs R.: What is the evidence that blood pressure monitoring is useful? In Problems of Neonatal Intensive Care Units. 59th Ross Conference on Pediatric Research, 1969, p. 81.

15. Laennec R. T. H.: A Treatise on the Diseases of the Chest. 1821.

16. Miller H. C., Behrle F. C.: Changing patterns of respiration in newborn infants. Pediatrics 12:141, 1953.

17. Miller H. C., Conkin E. V.: Clinical evaluation of respiratory insufficiency in newborn infants. Pediatrics 16:427, 1955.

18. Reid D. H. S., Tunstall M. E.: Recurrent neonatal apnea. Lancet ii: 155, 1965.

19. Wick H., Schmitt H.: Simple warning system for apnea in premature infants. Lancet i:880, 1967.

20. Stein I. M., Shannon D.: The pediatric pneumogram. A new method for detection and quantitating apnea. Pediatrics 55:599, 1975.

21. Smith S. C.: The heart in mothers and the newborn. J. A. M. A. LXXXIX-3, 1922.

22. Hess J. H., Lundeen E. C.: The Premature Infant, Its Medical and Nursing Care. Philadelphia: J. B. Lippincott Company, 1941.

23. Standards and Recommendations for Hospital Care of Newborn Infants. American Academy of Pediatrics, 1971.

24. Huch A., Huch R.: The development of the transcutaneous pO₂ technique into clinical tool. In Huch A., Huch R., Lucey J. F. (eds.): Continuous Blood Gas Monitoring. New York: March of Dime-Birth Defects Original Article Series, XV #4, 1.979, p. 5.

25. Eberhard P., Mindt W., Hammacher K.: Perkutane Messung des Sauerstat partal druckes Methodick and Anwendungen. Stuttgart Proc. Medizin Technik, 1972.

26. Bhat R., Vidyasagar D., Asonye U., Papazafiratou C.: Continuous tissue pH monitoring in critically ill neonates. J. Pediatr. 97:445, 1980.

27. Harris T. R., Nugent M.: Continuous arterial O₂ tension monitoring in the newborn infant. J. Pediatr. 82:929, 1973.

28. Yamanouchi I., Yamanouchi Y., Igarashi I.: Transcutaneous bilirubinometry. Preliminary studies of noninvasive transcutaneous bilirubinometer in the Okayama National Hospital. Pediatrics 65:195, 1980.

29. Heygi T., Hialt I. M., Indyk L.: Transcutaneous bilirubinometry. 1. Correlations in term infants. J. Pediatr. 98:454, 1981.

30. Lundberg N.: Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr. Scand. , 149 Supplement, 1960.

31. Corning J. L.: Spinal anesthesia in local medication of the cord. N. Y. Med. J. 42:483, 1865.

32. Quinke H.: Die lumbalpunctrian des hydrocephalus. Berl. Klin. W. Chnschr. 28:929, 1891.

33. Hayden P. W., Shurtleff D. B., Foltz E. L.: Ventricular fluid pressure recordings in hydrocephalic patients. Arch. Neurol. 23:147, 1970.

34. Tilbury M. S.: The intracranial pressure screw. A new assessment tool. Nuts. Clin. North Am. 9:641, 1974.

35. Numoto M., Slater J. P., Donaghy R. M. P.: An implantable switch for monitoring extradural pressure. J. Neurosurg. 42:249,1975.

36. Symon L., Dorsch N. W. C.: Use of long-term intracranial pressure measurement to assess hydrocephalic patients prior to shunt surgery. J. Neurosurg. 42:258, 1975.

37. Kindt G. W., Waldman J., Kohl S., et al.: Intracranial pressure in Reye's Syndrome. Monitoring and Control. J. A. M. A. 231:822, 1975.

38. Vert P., Andre M., Sibout M.: Continuous positive airway pressure and hydrocephalus. Lancet 2:319, 1973.

39. Davidoff L. M., Chamlin M.: The "Fontanometer" adaptation of the Schiotz Tonometer for the determination of intracranial pressure in the neonatal and early periods of infancy. Pediatrics 24:1065, 1959.

40. Purin V. R.: Measurement of the cerebrospinal fluid pressure in the infant without puncture. A new method. Pediatriia 43:82, 1964.

41. Wealthall S. R., Smallwood R.: Methods of measuring intracranial pressure via the fontanelle without puncture. J. Neurol. Neurosurg. Psychiatry 37:88, 1974.

42. Blaauw G., Van Der Bos J. L., Mus A.: On pulsations of the fontanelle. Dee. Med. Child. Neurol. 16:32:23, 1974.

43. Epstein F., Walk A., Hochwald G. M.: Intracranial pressure during compressive head wrapping in treatment of neonatal hydrocephalus. Pediatrics 54:786, 1974.

44. Vidyasagar D., Raju T. N. K., Chiang J.: Clinical significance of monitoring anterior fontanel pressure in sick neonates and infants. Pediatrics 62:996, 1978.

45. Krishnamoorthy K., et al.: Evaluation of neonatal intracranial hemorrhage by computerized tomography. Pediatrics 59:165, 1977.

46. Bada H. S., Hajjar W., Chua C., Sumner D.: Noninvasive diagnosis of neona-tal asphyxia and intraventricular hemorrhage by doppler ultrasound. J. Pediatr. 95:775, 1979.

47. Bejar R., et al.: Diagnosis and follow-up of intraventricular and intracerebral hemorrhage by ultrasound. Studies of infant's brain through the fontanelles and sutures. Pediatrics 66:661, 1980.

48. latrogenic problems in neonatal intensive care. 69th Ross Conference on Pediatric Research, 1976. Columbus, Ohio: Ross Company Publishers.

49. Estimated data from personal communication. Mead Johnson Nutritional Division and Ross Laboratories.

50. Sinclair J. C., Torrance W., Boyle M. H., et al.: Evaluation of neonatal intensive care programs. N. Engl. J. Med. 305:489, 1981.

51. Kitchen W. H., Campbell D. G.: Controlled trial of intensive care for very low birth weight infants. Pediatrics 48:711, 1971.

52. Kleinman J. C., Kovar M. G., Feldman J. J., Young C. A.: A comparison of 1960 and 1973-74 early neonatal mortality in selected states; Am. J. Epidemiol. 108:454, 1978.

53. Lee K. S., Paneth N., Gartner L. H., Pearlman M_. A., Gruss L.: Neonatal mortality: an analysis of the recent improvement in the United States. Am. J. Public Health 70:15, 1980.