Heroic experiments in man by Abraham Guz: breathing and not breathing

Membership

Mark Noble
Department of Medicine and Therapeutics, University of Aberdeen Medical School, UK

---

https://doi.org/10.36866/pn.99.42

Abraham Guz (1929–2014) was a respiratory physician who frequently observed the great distress in patients with dyspnoea – breathlessness – gasping for breath. He wanted to understand the mechanisms which mediated this very unpleasant sensation.

---

Prof Abraham Guz’s early work on lung reflexes was performed in London between 1964 and 1974. His death a year ago was followed by a memorial meeting on 5 November at Hodgkin Huxley House, under the kind auspices of the Physiological Society. Very little was said at the memorial meeting about these early studies, in which some heroic experiments on human volunteers were described in the days before ethical committees. This work surely deserves to
be better known.

Lung reflexes: How to get at the vagus nerves in humans

As the lungs are innervated by the vagus nerves, i.e., the X^th cranial nerve, it was natural to be interested in what afferent fibres in these nerves enter the brain stem and in the central connections giving rise to reflex effects and cerebral perceptions. By the 1960s, it had already been shown that inflation of the lungs in anaesthetised humans elicited a much weaker inhibition of breathing compared with animals, in whom the participation of the afferent vagus was proved. In 1963, Guz, with his colleagues Diana Trenchard and Mark Noble first tried to establish whether the brief stop in breathing in humans, induced by lung inflation was indeed mediated by the vagus nerves (the Hering Breuer inflation reflex). For this he needed a surgeon willing to allow him to put local anaesthetic on the nerves with the patients’ consent.

Transporting a laboratory to another hospital’s operating theatre

The willing surgeon worked at Charing Cross Hospital in the Strand, so the experiments involved much practical and logistic difficulty. Guz had his laboratory in the grounds of the Fulham Hospital near Hammersmith Broadway, but the patients undergoing thyroidectomy had their operations performed at Charing Cross Hospital in the Strand. Therefore his team had to transport all their recording equipment and Douglas bags (for the pressurised inflating air), etc., to the Strand in central London. There they met the theatre sister who looked askance at the cleanliness of this stuff and started damp-dusting it. For some of it, this was not a source of anxiety, but the recording equipment consisted of large cabinets filled with thermionic valves and cabinets containing moving coil galvanometers and photographic recording paper. Guz was worried about water getting into this electronic equipment and also about jolting of the delicate parts, particularly the thermionic valves and moving coil galvanometers.

The weak Hering-Breuer lung inflation reflex in man

The inflation reflex was indeed abolished by local anaesthesia on the vagi. The experiment was repeated using intra-bronchial intubation with a Carlens tube. That established that there was no cross over of signal from one lung through the contralateral vagus nerve.

Then Guz asked, ‘How can I study the individual afferent fibres?’

Abe’s collaborator John Widdicombe could do that in animals by shredding the nerve and recording directly from single fibres in the cut peripheral end. Even Guz would not have considered doing this in humans! He required a recording method that did not damage the nerve in any way; the antidromic method could fulfill this requirement, but could only explore afferent activity in large myelinated fibres that signal lung volume. It could not be used in thyroidectomy patients as a length of nerve was required, but was possible in radical operations to remove cancerous tissue. The results were very simlar to those obtained in dogs and cats, suggesting that there is a central nervous loss of the lung inflation reflex sensitivity in humans compared with animals.

Then Guz said, ‘I want to deflate lungs as well’

In humans study of the deflation reflex, which accelerates breathing, was beset by methodological difficulties until it was attempted by Guz by allowing a lung to collapse in patients being treated for pneumothorax. That treatment in 1971 was by inserting a tube into the air filled pleural space and putting the other end of the tube under water in a bedside bottle (sometimes negative pressure was applied to the bottle). When this tube was opened, air went back into the pleural space and the lung collapsed. This was accompanied by an immediate increase in breathing rate and electrical activity recorded in inspiratory muscles. This confirmed that humans did have a deflation reflex, but uncertainty persists in the absence of a repeat of the experiment with vagal block, and it is difficult to be sure how the strength of the reflex in humans compares with other species. In the 1960s and seventies, the Guz group also carried out a series of animal experiments to probe these and other lung reflexes.

Perception of events in the lungs and breathing: breath-holding

Paradoxically, Guz’s first investigation of perception of breathing was to experiment with breath holding! He persuaded his colleagues Mark Noble (MN) and John Widdicombe (JW) to undergo vagal and glossophayngeal block with local anaesthetic injected at the base of the skull (having persuaded an anaesthetist to put needles into the right place!). The pattern and sensation of normal breathing was unchanged in both subjects. The first subject (MN) held his breath after breathing air. Breath holding time at total lung capacity was prolonged from 100 seconds control to 230 seconds during block; so long indeed that he became cyanosed. Therefore, in the second subject (JW), the experiment was performed with him breathing 100% oxygen. His breath holding time at total lung capacity increased from 110 seconds to 215 seconds. Thus, there was a clear diminution in the unpleasant sensation that increases during breath holding until one is obliged to take a breath. But was this due to block of the vagus or glossopharyngeal nerves (containing afferent pathways from chemoreceptors)? We plumped for attribution to the vagus nerves because the first breath after breatholding gives relief even if the inspired air is a mixture of nitrogen and carbon dioxide.

Not breathing due to paralysis

An alternative approach was then taken by Campbell, who paralysed himself with curare and found that he could stand lack of inflation from a ventilator for much longer than he could hold his breath under normal conditions. His hypothesis was that respiratory sensations arose from the muscles that produce breathing. This idea arose from the results of Agostoni, who recorded increasing frequency and intensity of activity in the diaphragm, (recorded by electromyogram) during breath holding. Campbell then persuaded Mark Noble to be his second subject because he had also been a subject for the vagal block experiments of Guz. The result was the same in MN as it was in Campbell. The idea that diaphragmatic contractions were involved in breath holding sensation seemed to be confirmed by the results of phrenic nerve block by the Guz group, and by the absence of the sensation in a patient with diaphragm paralysed by poliomyelitis. There was then an attempt to find if muscles other than the diaphragm were a source for this sensation. The intercostal muscles seemed not to be involved because the sensation was still present during block of these muscles. This was compatible with studies in patients with mid or low cervical spinal cord transection, who had normal breath holding sensation. However, a patient with a C3 transection maintained his breathing with his sternomastoid muscles only; this patient had no breath holding sensation. One needs an intact vagus nerve and a contracting diaphragm to ellicit this sensation.

Not a travelling laboratory full of metal barrels please!

Perhaps this concentration of studies on breath holding is not relevant to that which is sensed during breathing and breathlessness. Other sensations that were intensively studied during these years were the detection of added elastic loads and the detection of added resistive loads to breathing. These probably share the same mechanism as both cause sub-atmospheric airway pressure during inspiration. Elastic loads consist of metal barrels which are switched into the patients airway with a Douglas tap. The wide range of elastic loads of varying magnitude and thus different sized barrels required Guz to fill the operating theatre at Fulham with all these barrels, when he was doing vagal plus pharyngeal nerve block at the base of the skull (more extensive damp dusting required). The ability to detect added elastic loads was not impaired by these blocks. Guz then had the good idea of doing this test on patients with cervical cord transections, but this necessitated travelling all the way to Stoke Mandeville Hospital in the Chilterns. Even he could not contemplate transporting a lorry load of metal barrels all that way, so he switched to resistive loads, which are a set of tubes of varying linear resistance. The ability to detect these added resistive loads was not affected by cervical transection at C3 level, thus excluding the possibility that the sensation was mediated by chest wall receptors. In patients with tracheostomies, resistive loads could be detected, but became more sensitive if the tracheostomy cuff was deflated and the ability detect added resistive load is
impaired by local anaesthesia of the mouth. The ability to detect added loads in subjects with vagus nerve blockade could therefore have been mediated by receptors in the upper airways, such as afferents running in the laryngeal nerve.

Guz, ‘I’ll make my friend John Widdicombe breathless with asphyxia’

Guz was primarily interested in breathlessness, but the studies described above contribute little to the understanding of this sensation. The loads used in the foregoing paragraph are very low and not unpleasant. Breathing through much higher resistances is required to cause unpleasant breathing. Breathing CO₂ causes hyperpnoea that becomes unpleasant at high minute ventilation. This was studied in the period of interest in the subject JW, using 7% CO₂ 93% O₂ rebreathing, with blocked vagus and glossopharyngeal nerves at the base of the skull. Peripheral chemoreceptor block was confirmed by the abolition of the ventilatory response to hypoxia, and hyperoxia inhibits the peripheral chemoreceptor response in any case. So it was the central drive from hypercapnia that was recorded. The lower ventilatory response to CO₂ after block, compared with control, was due to the loss of the peripheral chemoreceptor response to CO₂. The subject continued to breathe the hypercapnic inspired gas to a much higher end-tidal PCO₂, because the distressful sensation was absent. Guz stopped the experiment at a PCO₂ of 70mmHg for safety reasons. The authors concluded that the results were compatible with the hypothesis that the loss of unpleasantness was related to loss of lung afferent activity in the vagus nerves. However, it has to be said that the actual peak minute ventilation was lower during block than during control, so that it is possible that the lower level of unpleasant respiratory sensation was related to the lower level of hyperneoa.

Partial paralysis

Another type of experimental respiratory distress was expressed by MN after the curare experiment. Prior to complete muscular paralysis, as the curare concentration was gradually increased, there was a phase of partial curarisation when the vital capacity became near to and below tidal volume. This caused considerable distress that was relieved by switching on intermittent positive pressure ventilation. It appears that respiratory distress induced in normal humans could be mediated by lung afferent or chest wall afferent neural activity until someone does the partial curarisation experiment during vagal block!

Ethical experiments with no ethical committees

The type of heroic experiments on man described above became very difficult or impossible with the advent of ethical committees. I consider that these experiments were ethical in that the subjects consented to them and came by no harm from them. I prefer to trust the scientist rather than a committee. The later work of Guz and his expanding team diversified into many aspects of pulmonary pathophysiology, but he did show that lung inflation is sensed in the cerebrum. However, without vagal, phrenic and chest wall blocks, it is not possible to be sure of the afferent pathway, although Guz postulated that it resided in lung receptors and vagal afferent fibres. Of the many studies by Guz after 1974, which are outside the scope of this article, the enthusiasm to understand respiratory sensation was his continued main concern leading to identification of the area of the cerebral cortex associated with inspiration and other breathing sensations.

Further reading

Widdicombe JG (2006). Reflexes from the lungs and airways: historical perspective. J Appl Physiol 101, 628-634

Parkes MJ (2005). Breath-holding and its breakpoint. Exp Physiol 91, 1-15