Stress and ectopic beats in ships' pilots

Journalof Psychosomalic Printed in Great Britain.

Research. Vol. 26, No. 6. pp. 559-569, 1982.

0023.3999/82/060559-l Pergamon

I $03.00/O Press Ltd.

STRESS AND ECTOPIC BEATS IN SHIPS’ PILOTS TREVOR C. COOK* and PETER M. M. CASHMANj1981; accepted in revisedform 3 March 1982)

(Received 26 October

Abstract-There has been much discussion about the significance and causes of ectopic beats. They may be found in the electrocardiograms of apparently healthy subjects, although ectopic beats with aberrant contour are commonly regarded as indicating an underlying cardiac pathology. There has been speculation that stress may play a part in precipitating them, but this does not necessarily imply that aberrant beats are a completely normal response. Ships’ pilots were studied at work using long-term ambulatory ECG recordings which were analysed for aberrant beats both visually and by computer. A group of eight subjects whose recordings contained more than one or two isolated aberrant beats also tended to have more indications of abnormality in their clinical electrocardiograms @ < 0.02), and to score higher on the Neuroticism scale of the Eysenck Personality inventory @ < 0.05), than a Control Group of six other pilots. At work, the aberrant beats occurred most often when the pilots were involved in demanding activities, notably manoeuvring ships in hazardous situations; these periods were frequently marked by large increase in heart rate. Coronary risk factors, such as serum cholesterol and overweight, while being generally high, did not discriminate between those subjects who produced frequent aberrant beats and those who did not. VERY HIGH heart rates have been observed in ships’ pilots during manoeuvres such as berthing and docking [l, 21. The heart rates were obtained from long-term ambulatory recordings of the ECG, and the high peaks are similar to those reported among aircraft pilots during analogous operations, notably during take-off and landing [3]. However, much ship-pilotage involves prolonged vigilance rather than concentrated “stress” [4]. The varying types of pilotage demand are broadly reflected in the heart rate of the pilots. In addition to heart rate variations, the continuous ambulatory ECG reveals other important cardiac phenomena, such as the occurrence of ectopic beats. There is a considerable literature dealing with ectopic beats and their prognostic significance, both in cardiac disease and in the asymptomatic normal population, together with some speculation that environmental influences and the demands of work may play a part in the genesis of such arrhythmias [5, 61. In most of these studies, however, the psychological factors and the work circumstances have not been analysed in detail. The purpose of the present study is to try to identify any differences in psychological or physical characteristics and work stress between those pilots who produced ectopic beats during long-term recording and those who did not.

SUBJECTS

AND METHODS

The subject data and ambulatory ECG recordings were taken from the Department of Industry Human Factors in Marine Pilotage Project [l]. Subjects were initially selected on the basis of evidence of numbers of aberrant beats in their ECG recordings during rapid scanning of some fifty tapes. Cases were discarded if medical, psychological and shipboard workload data were not available or if the recordings contained excessive artefact which would compromise the analysis. Eight “Aberrant beat” subjects resulted when careful analysis of the recordings (by two methods which will be described) confirmed that they met the criterion of at least one aberrant beat per hour on average. *Occupational Psychology Department, Birkbeck College, London University and Bioengineering, Clinical Research Centre, Harrow, HAI 3UJ, Middlesex. tDivision of Bioengineering, Clinical Research Centre, Harrow HA1 3UJ, Middlesex. 559

Division

of

560

TREVOR C. COOK and PETER M. M. CASHMAN

The Control subjects were required to have clear recorded ECG’s of at least 11 h duration, to have medical, psychological and shipboard data available, and not to be taking any form of medication. Six Control subjects emerged from this allocation, and the analysis methods confirmed that three of these men had no aberrant beats in their tapes while the others only had them on isolated occasions (less than one per five hours). Median tape duration was 1030 min (range 240-I 100 min) for the AB group; and 1265 min (range 660-2090 mitt) for the Controls; all tapes included periods of piloting and ‘other activities’. On average, 67% of the total recording time (14705 mitt) covered ‘other activities’, 28% was pilotage from the bridge, and the remaining 5% was boarding and disembarking in approximately equal proportions. The median time spent on the bridge was 4% hours (which covered two jobs in some cases) and there was no significant difference between the bridge piloting durations for the two groups. The overall tape durations were significantly longer for the control than for the AB group (one-tailed p < 0.01, Mann-Whitney Test). The workload, psychological and medical data were compared for the two groups. Workload records were used to identify the following periods: non-pilotage activities; boarding and disembarking (usually involving substantial physical exertion); piloting from the bridge. Psychological characteristics were expressed as Extraversion and Neuroticism scores from Form B of the Eysenck Personality Inventory [7]. Medical data included age, systolic blood pressure, serum cholesterol, tobacco and alcohol consumption and parental mortality. The Quetelet height-weight index was calculated as Q = weight (kg) x 10 + (height(m))* [8]. Twelve-lead clinical ECG’s were carefully examined for noteworthy features according to the following criteria. For the resting ECG, it was decided to note any abnormality which would qualify for a coding under the Minnesota Code [9] plus some additional signs: 7’(V,) higher than T(V6); ventricular ectopic beat followed by a post-ectopic T-wave change; QT interval more than half the RR interval; left axis deviation between 0” and ~30” [lo]. For the post-exercise ECG’s the Minnesota Code was not used, and any STJ depression with upward-sloping ST segment, not persisting beyond two minutes, was disregarded. ST segment depression with horizontality was counted as a significant feature, as was any of the abovementioned ‘additional signs’. The ECG’s were read initially by one of the authors (TCC), and then re-read by a consultant cardiologist who also noted which records he would regard as clinically abnormal. The cardiologist was provided with the list of signs considered to be present or absent in each ECG but no other data.

Limitations of the ambulatory ECG recordings Because most of the long-term ECG recordings were obtained using only a single bipolar chest lead (approximately CM5), it would have been unrealistic to attempt to distinguish true ventricular ectopic beats from aberrantly-conducted supraventricular ectopics. For the purpose of the study, the term ‘Aberrant beat’ is therefore used to apply to both of these categories. Additionally, because atria1 ectopics and supraventricular ectopics with normal conduction were not reliably detected by the two methods of analysis used, they were rejected as a basis for classification. The standard Medilog 4-24 recorders used to collect the data did not accurately register signal components below 0.5 Hz, and ST segment changes could not be quantitatively examined. (Since this study began, Balasubramanian [l l] has reported on the superior ST segment registration of the newer Medilog II recorder.)

Analysis of the ECG recordings Two independent analyses were carried out. Firstly, a predominantly visual rhythm scan; secondly, a computer classification based on QRS morphology, using the Clinical Research Centre’s GRETA system [12]. By the first method, ectopic beats could be detected visually on a monitor oscilloscope, as spikes on a chart recorder write-out of the beat-to-beat pulse interval time (P.I.T.), or by having a prototype Medilog ECG Analyser (Oxford Medical Systems Ltd.) reproduce suspect episodes on standard ECG paper. All three techniques were used for each tape, replaying at sixty times real time. The GRETA system, however, classifies QRS complexes automatically but adapts to gradual changes as the tape progresses; it was set to run out samples of ECG whenever a new morphology was encountered, and the operator could intervene to correct misclassifications, by observation of ‘templates’ continuously displayed, or from the printed samples.

RESULTS 1.

Comparison The average

AND COMMENTS

of aberrant beat rates obtained by two methods of analysis aberrant beat (AB) rates obtained from the “Aberrant beat”

group

Stress and ectopic beats in ships’ pilots

561

ranged from 0.025 per minute to 0.665 per minute according to Greta, and from 0.034 to 0.649 per minute by the visual method. In the case of the six Control subjects, the agreement was virtually 100%: three men had nil AB rates, and the other three had only isolated AB’s on rare occasions. Overall, taking both groups together, the Spearman’s rank-order correlation coefficient for AB rates by visual and computer methods was 0.99, and GRETA detected more AB’s than the visual scan (Wilcoxon test: two-tailed p = 0.05), although five of the fourteen pairs of results were ties. The quality of recordings was very good, but it was impossible to avoid artefactual ‘noise’ entirely. GRETA was found to be resistant to false counting and was capable of detecting AB’s buried in substantial noise, while the visual scan operator (TCC) felt that he was missing abnormalities during bouts of frequent AB’s, especially if these occurred together with artefact. On the other hand, the visual scan could be more effective than GRETA. Figure 1 shows a ventricular tachycardia which GRETA misclassified but which was identified in the visual scan.

FIG. 1.-A

run of ventricular

tachycardia (Subject F) which was detected as artefact by the GRETA system.

by visual scan but classified

2. Comparisons between the aberrant beat group and the control group Ideally, one would have wanted all the tapes to be of similar length, but duration was dictated more by shipping conditions than by research requirements. The shortest tape (240 min) included only about one-and-a-half hours of piloting, while the longer tapes usually included much longer periods of time spent on the bridges of ships. Furthermore, it is most unlikely that all the acts of pilotage were of similar difficulty. It is also impossible to be sure than none of the Controls would have produced numbers of aberrant beats, given unlimited recording time; this kind of uncertainty is inherent in any study of long-term ECG recording [13, 141. The clinical variables and specific risk factors for the two groups are shown in Tables I and II. It is evident that the EPI Form B Neuroticism scores [7] and the 12-lead ECG codings are the only features which discriminate between the groups, with the overall heart rate from the taped ECG’s being of borderline significance. The Quetelet Indices and serum cholesterol levels are on the high side in both groups; an explanation may be found in the fact that these men are of high socio-economic status with little demand for physical exercise-a life-style which originated in their previous careers as merchant navy officers. Table II shows the most interesting variables for the individual subjects. While it was not the aim of this study to make diagnoses of illness, it is clear that the suspect clinical ECG’s tended to occur in the Aberrant Beat group rather than in the Control group, and that one or two of these men were not at all healthy cardio-

562

and PETERM.M.CASHMAN

TREVORC.COOK

TABLE I.--UPPER TABLE: MANN-WHITNEY UOR STUDENT'S ~TESTS FOR AGE,MEAN AMBULATORY HEART RATE, QUETELET INDEX, SYSTOLIC BLOOD PRESSURE, SERUM CHOLESTEROL LEVEL, EXTRAVERSION AND NEUROTICISM SCORES BETWEEN ABERRANT BEAT AND CONTROL GROUPS OF EIGHT AND SIX SUBJECTS RESPECTIVELY. LOWER TABLE:PROPORTIONS OF THE SAME GROUPS HAVING HYPOTHETICAL LIFE INSURANCE LOADINoS,BEINo NON-SMOKERS,DRINKINoALCOHOL ONLYOCCASIONALLY OR NOT AT ALL,HAVlNG A PARENT DECEASEDATAN AGE BELOW 65 YEARS,HAVINGACLINICAL ECGFEATURE NOTED(ASLISTED IN TABLE III), ANDHAVING

ACLINICALLYABNORMAL

ECG

SBP (mmHg)

Chol (mmol/l)

E

N

AB Group

Mean/Median SD/Range

Med: 5 1.5 39-63

88.1 8.6

Med: 261 187-347

Med: 135 120-200

7.5 1.3

11.4 4.7

12.6 3.2

Control Group

Mean/Median SD/Range

Med: 46.0 40-50

78.3 9.4

Med: 266 244-288

Med: 135 120-160

7.4 1.0

13.5 3.1

7.3 3.8

t=0.89

t=2.61

Statistic P

v= 13.5

t=1.9

0.10 (l-tail)

0.08 (2.tail)

With insur. load

Non-smoker

iJ=24

u=22

NS

NS

t=0.09 NS

NS

0.025 (2-tail)

Alcohol: none or occasional

Parent deceased < 65

With noted ECG feature

With ECG abnormal

0.88

0.50

AB Group

Proportion

0.50

0.50

0.37

0.63

Control Group

Proportion

0.33

0.67

0.67

0.33

0.17

0.17

NS

NS

NS

NS

0.016

NS

p (Fisher Test)

logically. Subject E, for instance, appears to be a “high risk” individual for coronary heart disease; he also recorded the highest AB rate. For convenience of reference, the aberrant beat rates obtained by the two methods of analysis have been averaged for each subject. 3. Heart rates, aberrant beat rates and workload Apart from peaks in the heart rate during the physical exertion of boarding and disembarking from ships, increases in heart rate were typically associated with manoeuvring a ship close to a source of danger. The situation where two ships may collide is one example, fortunately rather rare. Far more common is the danger of going aground, of hitting a jetty or quay during berthing. There is, of course, a risk of striking another vessel when anchoring in the official anchorages, and, in some circumstances, it may be difficult to determine from a confusion of lights and reflections whether other vessels are anchored or under way. Table III lists the one-minute peak values of heart rate for the fourteen cases, and indicates the activity in pilotage which was associated with each peak. Although the Aberrant Beat subjects tended to have higher mean heart rates than the Control subjects (Table I), there was no significant difference between the two groups in the percentage increases of peak heart rate above the ‘Other activities’ level. There is a tendency for higher heart rates to accompany higher aberrant beat rates in these subjects. Mean heart rate correlates with aberrant beat rate (Spearman’s rho = 0.52) for ‘Other activities’, and the Aberrant Beat group was found to have

0.002 0.000 0.000 0.000 0.001

0.002

93 68 85 83 67

74

1 J K L Mt

NT

0.035 0.487 0.094

78 89 88

0.657

given.

284

247 256 276 244 288

234 259 187

347

270

0.067

F G H

81

D

299

258 263

Q

0.035

97

17

C-t

0.392 0.036

Et

103 93

At Bt

AB rate

*NAD = No abnormality detected. j-Hypothetical life insurance loading

Control Group

AB Group

HR 13 5

8.6 1.6 7.5 8.5

130 150 140 120

13

14 I4 12

6 2 8 11 4

13

8.0

6.2 8.6 4.7

7.2 6.1 8.5 7.0 9.0

6.6

200

130 130 140

140 130 140 130 120

160

17

13

N

Chol

SBP

NAD NAD NAD NAD tall T, 11 mm not meeting criterion for Minnesota Code 9-5; NAD

NAD

NAD NAD NAD NAD Borderline

(V5)

ST dep. (11, VS)

NAD NAD ST dep. 1.25 mm with horizontally, persisting

NAD ST dep. (VS), with horizontally, persisting at at 3 min; PVC with postextra-systolic Twave change; QT more than half RR, most traces No exercise ECG

;

Post-exercise NAD STJ dep. 1 mm + persisting (V5)

Clinical ECG

R(V5)+S(V1)>35mm. ST dep. 1 mm horizontal or downward sloping (I, AVL, V5,6,7); Tnegative 1 min + (I, V4, 5,697) R( V5) = 27 mm R(V5)+S(V1)=36mm NAD

NAD* T(Vl)> T(V6); STJ dep’ (V4), 0.75 mm QRS axis - 20” Low voltage (I, II, III) 7-( Vl) > T( V6);

Rest

LEVELS,

Normal

Normal Normal Normal Normal Borderline

Normal Normal Abnormal

Abnormal

Normal Abnormal

Normal Abnormal

Opinion

AMBULATORY HEART RATESAND ABERRANTBEATRATES, QUETELET INDICES, SYSTOLICBLOODPRESSURES,SERUM CHOLESTEROL ScORES,S"MMARYOFFEAT"RESNO~EDINICAL ECG’s ANDCARDIOLOGIST'SOPINIONOFTHE ECG

NEUROTlClSM

S

TABLE II.-MEAN

TREVOR C. COOK and PETER M. M. CASHMAN

564 TABLE III.-HEART

RATESANDACTIVITY.

‘OTHER ACTIVITIES'COVERS ALL PERIODS OTHER THAN BOARDING, AND PILOTAGE

DISEMBARKING

S A B C D

L M N

Other activities (Mean)

Boarding (Peak)

Disembarking (Peak)

Pilotage (Peak)

98 88 72 81

143 133 136 133

118 109 145 135

160 124 91 110

92 17 89 90

145 128 128 101

139 104 112 105

116 96 106 133

89 70 89

184 113 133

96 131

125 69 98

81 66 73

132 146 113

113 127 105

100 104 97

(activity) _ Berthing Entering dock Taking over ship Manoeuvring for launch ship As for D, above Anchoring Taking over ship Berthing

coming

to

Overtaking in river (Peak inconspicuous) Manoeuvring for launch coming to ship In narrow channel Anchoring Passing through narrow harbour entrance

generally higher overall mean heart rates than the Control group. On the other hand, there is the paradox that an increase in heart rate may effectively inhibit extrasystoles in a subject who is prone to them. Figure 2 illustrates such an episode from the recording of Subject A. Throughout a period of over six hours when piloting he had sustained a sinus tachycardia averaging 111 bpm, but produced runs of ectopic beats not only after the physical exercise of disembarking but also as the heart rate decreased following the non-physical stress of berthing the ship.

FIG. 2.-Twenty-minute sample of P.I.T. chart recording from Subject A who sustained an average heart rate of over 111 bpm during 6 h of pilotage, inhibiting aberrant beats which appear later as the heart rate falls.

Figure 3 shows patterns of aberrant beat production which are more typical of the AB group subjects. In these examples, it seems to be the psychological stress of pilotage which both raises the heart rate and triggers the appearance of aberrant beats. Table IV lists the activities and aberrant beat rates for ten-minute periods when the maximum recorded AB’s for each subject occurred. Although five control subjects never had more than one AB in a ten-minute period, and were therefore omitted from the table, it is evident that the majority of the maximal AB episodes were work-related, Subject G providing the exception.

Stress and ectopic beats in ships’ pilots

x

x

x

565

x

x

__-________berthing_____,___-________________________

---_-_-__--_-

X

x x

-------------------docking

____ -___ro~tin.____________-___--__

X

_____

-__-

xx

___________________________________

FIG. 3.-Twenty-minute samples of P.I.T. chart recording (Subjects H and B). In each case, top strip is middle strip from routine pilotage, and bottom strip berthing or docking. from ‘Other activities’, AB’s are marked ‘x’ and sample one-minute heart rates are given. (NB: each strip covers 20 min; 60 min recording for each subject .)

In Table V, the AB rates during boarding, disembarking and pilotage are compared with those during all ‘Other activities’. Only the results for the AB group are shown, because the Control group all produced AB rates close to zero, making statistical testing inappropriate.

TREVOR C COOK

566

and PETER M M CASHMAN

TABLE IV.-AB RATES (PER MINUTE) FOR TEN-MINUTE PERIODS WHEN THESE RATES WERE MAXIMAL, ON-GOING ACTIVITIES, APPROXIMATETIME OF DAY, AND MEAN HEART RATE (bpm) FOR THE TEN MINUTES

S

AB rate

A B C D E F

6.400 0.200 0.300 0.700 2.550

G H

1.450 0.500

I

0.200

1.050

Activity

Following disembarkation Docking After taking over ship Manoeuvring for launch Manoeuvring for launch Anchoring Asleep Eating snack (had been running after disembarking) Boarding

J, K, L, M and N: never recorded

TABLE V.-AB DURING

or occasion

DISEMBARKING

HR

1200 0900 2300 0930 1800 0700 0230 1530

100 104 87 96 104 87 68 87

1400

148

more than 1 AB in any IO-min period

RATES FOR THE ABERRANT BEAT

BOARDING,

for train

TOD

AND

BRIDGE

GROUP

(N=S),

PILOTAGE

COMPARING

WITH

THOSE

RATES DURING

‘OTHERACTIVITIES'.(AVERAGERATESFROMBOTHANALYSES)

‘Other activities’ Median Range Wilcoxon p (l-tail)

0.050 0.004-0.536 T

Boarding 0.206 0.008-0.650 12 NS

Disembarking 0.182 0.000~0.850 1 0.025

Pilotage 0.104 0.038-0.961 7 0.10

Whilst the data of Table V support the suggestion that aberrant beats are often associated with the stresses of pilotage, it has to be admitted that no very clear statistical significance of differences between the AB rates in boarding, disembarking and pilotage, compared to those in ‘Other activities’, had been obtained. Apart from the small size of the group, it is worth noting that Subject A’s pilotage tachycardia, discussed earlier, and his relatively low AB rate while it was occurring, can be held responsible for the only-marginal significance of differences between pilotage and ‘Other activities’; all the other Aberrant Beat group subjects had higher AB rates in pilotage than in ‘Other activities’, as did two of the Controls who produced AB’s on isolated occasions.

DISCUSSION

The mutually-confirmatory overall findings from two independent methods of analysis indicated clearly the ordering and grouping of the subjects as producers of aberrant beats, and where the AB’s tended to occur. However, this statement must not be allowed to hide the fact that the two methods disagreed quite widely at times, and neither could be regarded as giving completely accurate results, even when searching for complexes as bizarre as aberrant beats. The disagreements seem to be due to the human observer having difficulties when AB rates were high, and the computer having difficulties when a new morphology appeared for the first time

Stress and ectopic beats in ships’ pilots

561

in circumstances where it could be interpreted as ‘noise’. A more up-to-date Oxford Instrument Analyser than the prototype instrument used by the visual analyst could provide exact timing (by direct reading of crystal-controlled clock signals recorded on the tapes), and this would open up greater possibilities for reducing the discrepancies between the two methods. The fact remains that the human analyst was able to establish timings fairly well on the prototype analyser by using recorded event marks, and, for the PIT chart-recordings and accompanying oscilloscope ECG displays, the recorded clock pulses were available; he did not manage to spot all the AB’s, and nor did the computer. The task of searching for ectopic beats in ambulatory recordings should not be under-estimated. The dilemma of deciding whether the aberrant beats observed in a single ambulatory recording lead were ventricular or supraventricular was avoided, but it is virtually certain that many were ventricular. Hinkle, Carver and Stevens [15] took the line that ventricular premature contractions (PVC’s) and complex ventricular dysrhythmias, in generally asymptomatic and outwardly healthy men, were indicative of coronary heart disease. This was held to be especially true for subjects with PVC rates which would translate approximately to 0.833 or more per minute. None of the pilot-subjects in this study had an overall AB rate of 0.833 per minute, although some exceeded it in the short term. On the other hand, some authors have said repeatedly that ectopic beats are usually of no serious significance and that normal people may have them [16]. It can be readily agreed that ectopic beats occur very commonly in apparently healthy subjects, and evidence from a variety of other studies now supports this view [5, 14, 171. As far as aberrant, possibly ventricular, beats are concerned, the Control subjects are either totally free of them or have only one or two in many hours’ recording, while the AB group includes individuals with varying AB rates overall but who reveal phases of increased AB rates, notably at times of stress-physical or ‘mental’. All these subjects were normal, at least in the sense that they were at work in responsible jobs, and were most certainly not overtly having difficulty in coping with the challenges of daily life for health reasons. There is no doubt that ship pilotage can be stressful. The large heart rate increases in the absence of physical exertion seem to reflect exactly the cardiac-somatic decouping phenomenon described by Obrist [ 181. The increases are likely to be due to sympathetic nervous system activation, produced by the threat of consequences of false moves. This activation may precipitate also the appearance of ectopic beats; perhaps through the influence of catecholamines, increasing the irritability of the heart of soldiers” (Da Costa’s Syndrome, or neuromyocardium. The “irritable circulatory asthenia) seems to have possible relevance to the question [19], and there is an extensive scientific literature which supports popular and traditional beliefs that the heart is sensitive to what is in “the mind” [19, 20, 22, 23, 24, 251. Animal studies carried out half a century ago showed that stimulation of the hypothalamus produced ventricular ectopics, and implicated sympathetic pathways to the heart and paths controlling secretion of adrenaline [28]. It is consistent with the notion of “anxiety” giving rise to ectopics that the AB group in the present study scored significantly higher on Eysenck Personality Inventory Neuroticism than the Controls. The Framingham studies [26] found little evidence of psychological influence on

568

TREVOR C. COOK and PETER M. M. CASHMAN

the incidence of coronary heart disease; the coping with stress by different personalities, and evidence of anxiety, were said to be unrelated to the disease or its precursors. Framingham reports emphasized the importance of cholesterol levels in determining the risk of heart disease. The pilotage control subjects’ total serum cholesterol levels are as high as those of the AB group, and, taking Hinkle’s criterion [15] for moderate to high risk, all but two of the AB group, and all but one of the Controls, would qualify for inclusion as risks on this aspect of his scoring system. Blood pressure, another Framingham risk factor, does not discriminate between the AB group and the Controls; but having some reportable feature in the clinical ECG does discriminate, even if the feature was too “minor” to warrant being classed as definitely abnormal. It seems that psychological and environmental influences readily affect the heart, chiefly to alter its sinus rate, but also, especially in certain types of subjects, to induce aberrant beats. These are the two phenomena above all which have been observed in the pilotage subjects as their response to work stress. It is true that there was one subject who produced his maximum AB rate during sleep, but this is not unexpected as a minority finding [27]. Aberrant beats occurred most commonly at times when the heart rate was raised (or had been raised) in response to some environmental stress. The men who produced them tended to have higher neuroticism scores and features in the clinical ECG’s which (to varying degrees) have been held to be indicative of cardiovascular impairment, while those who rarely or never produced aberrant beats usually had lower neuroticism scores, and clinical ECG’s which were completely normal. The aberrant beats certainly did not occur randomly, and it is far more likely that they were dependent upon the individual’s personal responsivity to stress, the severity of the stress, and the organic state of his heart and blood vessels, than upon any single factor, such as a particular cardiac cycle length, per se, favouring the genesis of ectopic beats. The psychological influence on the incidence of aberrant beats and on the heart rate is clear enough, but no conclusions can be confidently made about the cause and effect relationships between aberrant beats and coronary heart disease; although signs of coronary insufficiency in the clinical ECG accompanied the occurrence of aberrant beats in the ambulatory recording of some subjects, other coronary ‘risk’ factors, perhaps most noticeably the cholesterol levels, were apparently quite independent of the AB rate. The indications are that physically or ‘mentally’ demanding situations are more likely to give rise to aberrant beats in subjects with higher levels of neuroticism and with notable features in their clinical electrocardiograms than in those with low neuroticism scores and unequivocally normal clinical electrocardiograms, although occasional aberrant beats may occur in healthy people. However, the numbers of subjects are small, and it is felt that the most appropriate recommendation which can be made is the one very commonly offered in small group research reports: there is scope for further study. Particularly, there is need to use more recent versions of the instruments employed in this study so as to be able to locate relevant events and cardiac phenomena very accurately in time, and it is strongly recommended that at least two ambulatory ECG leads should be recorded so that the origin and type of ectopic beat can be determined with some confidence. While this remark especially applies to the predominantly visual, human-operator-orientated method of analysis, it is also true that the GRETA system will benefit from having a second view of QRS morphology.

Stress and ectopic beats in ships’ pilots

569

Acknowledgements-We are grateful to Dr. J. H. Briggs, Harley Street, London, and to Dr. P. Shipley, Occupational Psychology Department, Birkbeck College, London University, for allowing the use of data from the Human Factors in Marine Pilotage Project; also to Dr. V. Balasubramanian, Consultant Cardiologist, Northwick Park Hospital, Harrow, who verified the ECG ratings and to Mrs. J. F. Asquith for the typing. REFERENCES 1. SHIPLEY P. A Human Facfors Sludy of Marine Pilotage. Birkbeck College, London University. Department of Industry, London, 1978. thesis, Birkbeck College, London 2. COOK TC. A study of pilotage in a U.K. oil port. M.Phil. University, 1977. 3. ROSCOE AH. Heart rate changes in test pilots. In The Study of Heart Rate Variability (Edited by KITNEY RI, ROMPELMAN0). Oxford: Clarendon Press, 1980. 4. COOK TC, SHIPLEYP. Human factor studies of the working hours of U.K. ships’ pilots. Appl Ergonom 1980; II: 85. hour ambulatory electrocardiography from urban and rural populations. 5! GOULDING L. Twenty-four In Proceedings of ISAM 1977 (Edited by STOUT FD, RAFTERY EB, SLEIGHT P, GOULDING L). London: Academic Press, 1978. electrocardiography in car 6. BAXTER PJ, WHITE WG, BARNES GM, CASHMAN PMM. Ambulatory workers. Brit JIndustMed 1978; 35: 99. London: University of 7. EYSENCK HJ, EYSENCK SBF. Manual of the Eysenck Personality Inventory. London Press, 1964. ou mesure des diffkrentes facultCs de I’homme (1870). Quoted by 8. QUETELET A. Anthropometrie KHOSLA T, LOWE CR. The Height and Weight of British Men. Lancet 1968; (i): 742. of ECG Minnesota Code. In Cardiovascular Survey Methods. 9. ROSE GA, BLACKBURNH. Classification Geneva: W.H.O., 1968. of Electrocardiology. Oxford: Blackwell, 1976. 10. SCHAMROTHL. Introduction ST segment 11. BALASUBRAMANIANV, LAHIRI A, GREEN HL, STOTT FD, RAFTERY EB. Ambulatory monitoring. Brit Heart J 1980; 44: 419, program for continuous ECG processing in accelerated time. 12. CASHMAN PMM. A pattern-recognition Comput Biomed Research 1978; II: 31 I. electrocardiographic records in patients 13. GOLDBERG AD, RAFTERY EB, CASHMAN PMM. Ambulatory with transient cerebral attacks or palpitation. Brit Med J 1975; 4: 569. 14. CLARKE JM, HAMER J, SHELTON JR, TAYLOR S, VENNING GR. The rhythm of the normal human heart. Lancer 1976; (ii): 503. of asymptomatic disturbances of cardiac 15. HINKLE LE. CARVER ST, STEVENS M. The frequency rhythm and conduction in middle-aged men. American J Cardioll969; 24: 629. and tachycardia. In Signs and Symptoms, 5th Edition (Edited by MACBRYDE 16. MASSIE E. Palpitation CM, BLACKLOR RW). Philadelphia: Lippincott Company, 1970. 17 VERBAAN CJ, POOL J, VAN WANROOY J. Incidence of cardiac arrhythmias in a presumed healthy population. In Proceedings of ZSAM 1977 (Edited by STOTT FD, RAFTERY EB, SLEIGHT P, GOULDINC L). London: Academic Press, 1978. interaction. Psychophysiology 1976; 13: 95. 18 OBRIST PA. The cardiovascular-behavioural characteristics of neurocirculatory 19 TZIVONI D, STERN Z, KEREN A, STERN S. Electrocardiographic asthenia during everyday activities. Brit Heart J 1980; 44: 426. emotions and 20. STEVENSON IP, DUNCAN CH, WOLF S, RIPLEY HS, WOLF HG. Life situations, extrasystoles. Psychosom Med 1949; 11: 257. of Cardiac Arrhythmias. London: Butterworths, 1970. 21. STOCK JPP. Diagnosis and Treatment approaches to the study of catecholamines and emotion. In 22. FRANKENHAEUSER M. Experimental Emotions-Their Parameters and Measurement (Edited by LEVI L). New York: Raven Press, 1977. 23. LYNCH JJ. The Broken Heart. New York: Basic Books, 1977. of heart rate variability under the ergonomic 24. LL~CZAKH, PHILIPP U, ROHMERT W. Decomposition aspects of stressor analysis. In The Study of Heart Rate Variability (Edited by KITNEY RI, ROMPELMAN 0). Oxford: Clarendon Press, 1980. in catecholamine and cortisol excretion patterns over 25. LL~NDBERG U, FORSMAN L. Consistency experimental conditions. Pharmacol Biochem Behav 1980; 12: 449. epidemiology from Framingham. Am J Cardiol 1976; 26. KANNEL WB. Some lessons in cardiovascular 37: 269. changes in cardiac 27. DOHNO S, LYNCH JJ, PASKEWITZ DA, GIMBEL KS, THOMAS SA. Sleep-waking arrhythmia in a coronary care patient. Psychosom Med 1977; 39: 39. and anatomical evidence for the existence of nerve 28. BEATTIE J, BROW GR, LONG CNH. Physiological tracts connecting the hypothalamus with spinal sympathetic centres. Proc R Sot 1930; B 744: 253.