Journal of the Neurological Sciences 178 (2000) 66–69
www.elsevier.com / locate / jns

Neurological accidents caused by repetitive breath-hold dives:
two case reports
a,

c

b

Kiyotaka Kohshi *, Takahiko Katoh , Haruhiko Abe , Toshio Okudera
a

d

Department of Neurosurgery and Division of Hyperbaric Medicine, School of Medicine, University of Occupational and Environmental Health,
1 -1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807 -8555, Japan
b
Department of Internal Medicine, School of Medicine, University of Occupational and Environmental Health, 1 -1 Iseigaoka, Yahatanishi-ku,
Kitakyushu 807 -8555, Japan
c
Department of Health Information Science, School of Health Sciences, University of Occupational and Environmental Health, 1 -1 Iseigaoka,
Yahatanishi-ku, Kitakyushu 807 -8555, Japan
d
Department of Radiology, Akita Research Institute of Brain and Blood Vessels, 6 -10 Senshu-Kubota-machi, Akita 010 -0874, Japan
Received 17 February 2000; received in revised form 26 May 2000; accepted 26 June 2000

Abstract
We report two Japanese male professional breath-hold divers (33 and 39 years of age) who experienced neurological disorders during
repetitive dives to over 20 m of seawater. One patient had right homonymous hemianopsia, and the other presented with right hemiparesis
with facial involvement and sensory deficit. In addition, they each had a history of neurological problems following such dives. Magnetic
resonance images of their brains disclosed multiple T2-weighted hyperintensities corresponding to their neurological symptoms. Their
brain lesions suggest a multiple cerebral infarction caused by occlusion of the cerebral arteries. We conclude that the repetitive deep
breath-hold dives induced the brain involvement.  2000 Elsevier Science B.V. All rights reserved.
Keywords: Dysbaric accident; Decompression illness; Breath-hold diving; Breath-hold diver; Japanese ama diver; Cerebral infarction

1. Introduction
Dysbaric diving accidents, a collective term of all
pathological changes related to altered environmental
pressure, include two major forms: barotrauma and decompression sickness (DCS). Barotrauma involves tissue
damage resulting from a change in the volume of entrapped gas during decompression. Pulmonary barotrauma,
caused by alveolar gas expansion, liberates arterialized gas
into the systemic circulation, which leads to cerebral
arterial gas embolism. DCS occurs when inert gas bubbles
form (nitrogen in air breathing) during inadequate de*Corresponding author. Tel.: 181-93-603-1611; fax: 181-93-6914200.
E-mail address: k-kohshi@clnc.uoeh-u.ac.jp (K. Kohshi).

compression. The symptoms vary widely due to local
damage and intravascular blockage.
Dysbaric accidents in breath-hold divers are considered
very rare because of the lower permeability of alveolar
nitrogen gas into the circulation [1]. Although diving
accidents are known to have occurred among breath-hold
divers in the South Pacific [2], there have not been any
reported events among either Korean or Japanese professional breath-hold (ama) divers. Moreover, no such cases
have been reported for more than 3 decades. However, we
recently experienced 2 Japanese male ama divers with
histories of neurological disorders after repetitive dives [3].
More recently, we treated 2 other male ama divers who
experienced neurological problems during dives. We report
the characteristics of the neurological and neuroradiological findings caused by breath-hold dives.

0022-510X / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved.
PII: S0022-510X( 00 )00360-9

K. Kohshi et al. / Journal of the Neurological Sciences 178 (2000) 66 – 69

67

2. Case reports

2.1. Case 1
A 33-year-old man was referred to a local hospital on
May 27, 1996 due to visual disturbance in the right field.
He was a male ama diver on Mishima Island in Yamaguchi
Prefecture who started his profession at the age of 15. On
this island, every ama diver wears a full wet suit and
carries a weight belt equivalent to neutral buoyancy (7|8
kg). The patient was a partially assisted breath-hold diver
who descended to 15|25 m of sea water (msw; measured
by a fish finder) with a 15-kg weight and then ascended
without assistance. He engaged in daily diving work
during the harvesting season. During that time, he repeatedly made 1|1.5-min dives with 1-min surfacing
intervals between dives. He dove generally two shifts a
day, one in the morning and the other in the afternoon,
taking a lunch break in between. He would normally spend
5 to 6 h in the sea. On May 24, he began diving at 9:20
A.M. and continued diving to 22 msw until noon. After a
20-min lunch break, he began diving again. However, a
sudden onset of dizziness and blurred vision in the right
field occurred at 2:10 P.M. Three years before, he experienced transient salivation from the right corner of his
mouth after such dives. He smoked a pack of cigarettes per
day for 14 years, but he had no history of stroke or other
vascular diseases. An examination of the visual fields
showed right homonymous hemianopsia. Other neurological and physical examinations were normal. A full blood
count, standard biological examinations, chest radiograph,
12-lead electrocardiography and transesophageal echocardiography were normal. Magnetic resonance images (MRI)
of the brain on the 4th day after the onset showed two
hyperintense cerebral lesions in the left occipital lobe and
right basal ganglia (Fig. 1). The patient was treated with
dexamethasone (12 mg i.v. daily for 3 days), followed by
drug tapering of 4 mg daily over 3 days. His visual
disturbance regressed within 3 weeks but he had residual
right lower quadrantanopsia. MR angiography taken 7 days
later showed no evidence of a stenotic or obstructive
region in the cerebral and cervical arteries.

2.2. Case 2
A 39-year-old male ama diver was admitted to the same
hospital on October 4,1997 with moderate right
hemiparesis and a hemisensory disturbance. He lived on
Ainoshima Island near Mishima Island. His diving life and
patterns were almost the same as Case 1. On October 2,
1997, he began diving around 9:00 A.M. and continued to
dive between 20 and 23 msw. He noticed slight dizziness
during dives before noon, so he stopped diving and his
symptoms disappeared about 30 min later. After a 1-h
lunch break, he began diving again until he noticed motor
weakness and sensory numbness in his right side at 2:10

Fig. 1. MRI of the brain of Case 1. T2-weighted images (TR / TE:
2000 / 112) obtained on the 4th day after the accident show two increased
signal intensities in the left occipital lobe and the right basal ganglia.

68

K. Kohshi et al. / Journal of the Neurological Sciences 178 (2000) 66 – 69

P.M. At the ages of 17, 25 and 27 years, he suffered from
transient left hemiparesis during repetitive dives. Except
for these events, he had no history of stroke or vascular
disease. He had smoked half a pack of cigarettes per day
since the age of 19 years. The patient was transferred to
the hospital because his neurological symptoms progressed. Neurological examinations revealed moderate
right hemiparesis with facial involvement and a hemisensory disturbance. An MRI of the brain on the 3rd day after
the onset showed 3 hyperintense cerebral lesions (Fig. 2).
Other examinations including cerebral MR angiography, as
in Case 1, showed no abnormalities. He continued to
receive daily hyperbaric oxygen treatments (2.5 atmospheres absolute, 75 min) for 4 days, during which time his
neurological functions improved. A follow-up brain MRI
obtained 2 weeks later showed a reduction in the size of
the hyperintensities in the left parietal and basal ganglia,
except for the lesion in the right frontal lobe. He was
discharged with slight residual numbness in his right upper
limb.

3. Discussion

Fig. 2. MRI of the brain of Case 2. T2-weighted images (TR / TE:
2000 / 112) obtained on the 3rd day after the accident show three
increased signal intensities in the left parietal lobe and basal ganglia and
in the right frontal lobe.

Because these divers had neither vascular diseases nor
risk factors for stroke, except for smoking, and the events
happened during repetitive deep dives, there is a great
possibility that the breath-hold dives caused the brain
damage. The brain lesions demonstrated by MRI suggest a
vascular pathogenesis, especially occlusion of cerebral
arteries.
Diving accidents were reported in professional breathhold divers of the Tuamotu Archipelago in the South
Pacific [2]. The accidents were called ‘‘Taravana’’ diving
syndrome [2], which includes vertigo, nausea, partial or
complete paralysis, unconsciousness, and even sudden
death. Motor paralysis was the most common symptom
and was transient. Many diving physiologists suspect that
these disorders are compatible with DCS in the brain. In
our patients, the findings of neurological disorders were
close to the observations in ‘‘Taravana’’ diving syndrome.
Characteristically, these diving accidents in breath-hold
divers were limited to brain involvement. However, in
compressed air divers, the manifestations of neurological
DCS are commonly attributed to spinal cord damage [4].
The above symptoms were considered cerebral arterial gas
embolisms rather than DCS in the brain [4].
The mechanisms of brain damage following repetitive
breath-hold dives are poorly understood. Paulev [5] concluded that disorders caused by repetitive breath-hold dives
were DCS because they were immediately relieved by
recompression. Because of possible nitrogen accumulation
following repetitive deep breath-hold dives [5–7], nitrogen
gas bubbles might have formed in the intra- and extravascular spaces of the brain during or after decompres-

K. Kohshi et al. / Journal of the Neurological Sciences 178 (2000) 66 – 69

sion. Olszowka and Rahn, however, described that nitrogen accumulation in fatty tissues increases consecutively
throughout repetitive breath-hold dives despite quickly
reaching a steady state in the brain [7]. From the point of
nitrogen kinetics, it is difficult to explain how de novo
bubbles are generated in the brain and cerebral vessels. In
contrast, venous gas bubbles are caused by a continuous
release of nitrogen from peripheral fatty tissues during and
after decompression. By means of an ultrasonic bubble
detector, ‘‘silent’’ venous gas bubbles were recorded in a
Japanese ama diver after repetitive dives [8]. Since the
mammalian lungs usually constitute a competent filter for
venous bubbles larger than 21 mm in diameter [9,10],
smaller bubbles released by the lungs do not arrest in the
cerebral circulation [11]. Hills and James, however,
showed that such microbubbles transiently impair the
blood–brain barrier [12]. Microbubbles passing through
the lungs might have caused the brain involvement in the
ama divers, as suggested by Lanphier [6]. Moreover, other
possible risk factors or causes for the development of
diving accidents are smoking or bubble arterialization
through the lungs and / or via a right-to-left intracardiac
shunt [13–15]. Both smoking and an intracardiac shunt are
known to be risk factors for diving accidents in compressed air divers [14,15]. Although an intracardiac shunt
could not be demonstrated in our 2 patients, diving
accidents in breath-hold divers might be induced by these
risk factors or causes.
Diving accidents in Japanese ama divers have not been
previously reported. However, the diving patterns of these
ama divers are very similar to those of other places [16].
Why do other Japanese ama divers not develop clinical
signs of diving accidents? We may assume that many
diving accidents were transient and less serious than
‘‘Taravana’’ diving syndrome. Moreover, a possible explanation is the tendency of keeping diving accidents secret in
the Japanese ama communities.

Acknowledgements
We are grateful to Prof. A Yokota for careful review of
the manuscript.

69

References
[1] Hong SK, Rahn H, Kang DH, Song SH, Kang BS. Diving pattern,
lung volumes, and alveolar gas of the Korean diving women (ama).
J Appl Physiol 1963;18:457–65.
[2] Cross ER. Taravana diving syndrome in the Tuamotu diver. In:
Rahn E, Yokoyama T, editors. Physiology of breath-hold diving and
the ama of Japan. Washington, D.C.: Natl. Acad. Sci. — Natl. Res.
Council Publ. 1341; 1965, pp. 207–219.
[3] Kohshi K, Kinoshita Y, Abe H, Okudera T. Multiple cerebral
infarction in Japanese breath-hold divers: two case reports. Mt Sinai
J Med 1998;65:280–3.
[4] Melamed Y, Shupak A, Bitterman H. Medical problems associated
with underwater diving. N Engl J Med 1992;326:30–5.
[5] Paulev P. Decompression sickness following repeated breath-hold
dives. J Appl Physiol 1965;20:1028–31.
[6] Lanphier EH. Application of decompression tables to repeated
breath-hold dives. In: Rahn B, Yokoyama T, editors. Physiology of
breath-hold diving and the ama of Japan. Washington, DC: Natl.
Acad. Sci. — Natl. Res. Council Publ. 1341; 1965, pp. 227–236.
[7] Olszowka AJ, Rahn H. Gas store changes during repetitive breathhold diving. In: Shiraki K, Yousef MK, editors, Man in stressful
environments — diving, hyper-, and hypobaric physiology, Springfield, IL: Charles Thomas, 1987, pp. 41–56.
[8] Spencer MP, Okino H. Venous gas emboli following repeated breath
hold dives [abstract]. Fed Proc 1972;31:355.
[9] Butler BD, Hills BA. The lung as a filter for microbubbles. J Appl
Physiol 1979;47:537–43.
[10] Butler BD, Katz J. Vascular pressures and passage of gas emboli
through the pulmonary circulation. Undersea Biomed Res
1988;15:203–9.
[11] Heinemann HO, Fishman AP. Nonrespiratory functions of mammalian lung. Physiol Rev 1969;49:1–47.
[12] Hills BA, James PB. Microbubble damage to the blood–brain
barrier: relevance to decompression sickness. Undersea Biomed Res
1991;18:111–6.
[13] Hills BA, Butler BD. Migration of lung surfactant to pulmonary air
emboli. In: Bachrach AJ, Matzen MM, editors, Proceedings of the
7th Symposium on Underwater Physiology. Bethesda, MD: Undersea Medical Society; 1981, pp. 741–752.
[14] Wilmshurst PT, Byrne JC, Webb-Peploe MM. Relation between
interatrial shunts and decompression sickness. Lancet 1989;1:1302–
6.
[15] Wilmshurst P, Davidson C, O’Connell G, Byrne C. Role of
cardiorespiratory abnormalities, smoking and dive characteristics in
the manifestations of neurological decompression illness. Clin Sci
(Colch) 1994;86:297–303.
[16] Mohri M, Toni R, Nagaya K, Shiraki K, Elsner R, Takeuchi H et al.
Diving patterns of ama divers of Hegura Island. Japan Undersea
Hyperb Med 1995;22:137–43.