The word laser is an acronym for ‘Light Amplification by the Stimulated Emission of Radiation’. Some understanding of the nature of light is necessary to understand laser action. Light has a dual nature, behaving sometimes as a wave and at other times as a mechanical particle. The wave nature gives it wavelength; its particle side is manifested as the photon, which carries a discrete amount of energy. Both wave and particle viewpoints help towards an understanding of lasers but the particle theory is, in general, the more important.

Quantum mechanics limits atoms and molecules to certain discrete states of energy, the lowest of which is called the ground state. Atoms and molecules make instantaneous transitions between energy levels, which involve the absorption or release of energy. The transition energy is usually absorbed or released as a photon.

In order to understand the principle of a laser, consider a simple atom with orbiting electrons. The electrons orbiting the nucleus of the atom have specific energy levels; electrons away from the nucleus are in higher energy levels and electrons in the ground or lower energy states are closer to the nucleus. Now consider an incoming photon carrying a unit or quantum of energy that matches the difference between two energy levels of the atom. This photon can be absorbed, producing an excited atom in which an electron moves to a higher energy state; when this electron later spontaneously returns to the original lower energy state, the absorbed energy is released as another photon— ‘spontaneous emission’. There is another possibility, first predicted by Einstein in 1917: if the atom in the excited state is struck by another photon the energy is not absorbed, but it instead stimulates the atom to release a second photon of equal frequency and energy travelling in the same direction and in perfect spatial and temporal harmony with the stimulating photon (Fig. 1) 1416. This phenomenon is termed ‘stimulated emission of radiation’. Thus a photon may be absorbed or stimulate emission of another photon.

Left to themselves, atoms will always arrange themselves among their energy states in such a way that there are more in the lower energy states than in the higher states; in ordinary circumstances absorption always wins. Stimulated emission can, however, win over absorption, if for one pair of energy states there are more atoms in the higher than the lower state. Such a condition is called a population inversion, and is necessary for laser action.

Heating can increase the average energy of atoms or molecules, but heating alone cannot create a population inversion: heat moves all the atoms a little higher on the ladder, it does not increase the ratio of atoms in a higher state. Heat energy selectively excites atoms to certain high-energy states, where they keep their excess energy for an unusually long time. Atoms in a laser can be excited or pumped up to this state by light from other sources (even another laser), electrical currents passed through gases or semiconductors, by chemical reactions, or in other ways.

Stimulated emission produces light that travels in the same direction as the triggering light. In most lasers, however, stimulated emission is a weak effect that can build up to high power only after the light has travelled a long distance through the laser, interacting with other atoms. By placing a mirror at either end of the laser tube, the beam can be amplified to produce high powers. If one mirror is partially reflecting the laser beam can be allowed to emerge.

The light produced from lasers exhibits several special properties that differ in a number of respects from that of light from a domestic light bulb. Ordinary light is incoherent, whereas laser light is called coherent: the photons are travelling in the same direction and are identical, being in step with each other in both time and space. The coherence of laser light means that the divergence of a laser beam is very small (the beam is collimated). The full power of a laser beam can be focused on to a very small spot and it is this property of a laser that enables almost all of the power to be coupled into and transmitted through a small diameter fibreoptic guide which can in turn be passed through instrumentation channels of flexible endoscopes. Laser light consists of either one or just a few distinct colours. The beam of light from a laser is highly collimated, the irradiance (power per unit cross-sectional area) of the beam is very high, and can be focused to very tiny spots, producing enormous irradiance and localized power.

An understanding of the interactions of laser light with tissue is fundamental to the rational use of lasers as instruments in surgery. Interactions are best considered by examining the fate of a laser beam externally irradiating a block of tissue (Fig. 2) 1417. On striking an air/tissue interface some of the photons are reflected from the tissue surface. On entering tissue a photon may be scattered; its energy is not significantly altered but the direction of travel is changed. Light that has penetrated deeper than a few hundredths of a millimetre will have been subject to multiple scattering. It is important to note that scattered light does not produce any biological effect. Similarly, light that is transmitted through tissue exerts no biological effect: in order for a biological reaction to occur light must be absorbed by the tissue.

Non-specific absorption occurs when a variety of tissue agents absorb the light. In general absorption in tissue is determined by its constituents such as melanin, haemoglobin, myoglobin, and water. The absorption characteristics of individual tissues vary enormously and are also highly dependent on the wavelength of the light. The carbon dioxide laser beam (wavelength 10600 nm, in the far infrared/invisible end of the spectrum) is strongly absorbed by water, whereas the argon ion laser beam (blue/green) is strongly absorbed by haemoglobin molecules and other pigments. In certain circumstances it may be desirable to administer an exogenous agent, to produce specific absorption of light in certain tissues such as malignant tumours. This allows photochemical reactions to be produced and is the basis of photodynamic therapy.

Absorption of a photon in a non-specific manner may produce thermal changes in the tissue. These are at present the most widely used and surgically useful biological effects produced by a laser beam. If the rate of delivery of the photons (laser power) is such that the energy is dissipated in the surrounding medium as quickly as it is delivered no significant rise in the tissue temperature will occur. However, if the light is delivered at high enough power the tissue temperature will rise. Initially, at low rates, the temperature rise may be great enough to cause thermal damage to biological reactions. Local heating of malignant tissue to temperatures in the region of 41 to 45°C may produce selective hyperthermic destruction of malignant cells, since they are more sensitive to such temperatures than is normal tissue. This differential killing ability is lost at temperatures above 45°C and all cells are rapidly killed above 50°C. Further heating of tissue causes thermal contraction and coagulation of proteins: as the tissue shrinks small vessels can be sealed, arresting haemorrhage, and thrombosis of the occluded vessel seems to occur as a secondary event. This haemostatic effect of laser light is best, when the volume of tissue heated is relatively large (>5 mm). The Nd YAG (neodymium yttrium aluminium garnet) laser (wavelength 1064 nm, in the near infrared) is able to seal vessels up to 1 mm in diameter. If more energy is used, tissue necrosis occurs, with vaporization, laser ablation, and burning. Vaporization occurs at 100°C, when cellular water boils. Three lasers are used for their thermal effects: carbon dioxide (CO&sub2;), argon ion, and Nd YAG. The CO&sub2; laser beam is absorbed by water and so produces a very localized effect. The laser energy is rapidly absorbed by tissue, causing intracellular water to boil, disrupting cells, vaporizing, and cutting through the tissue. There is very little scattering of the CO&sub2; beam, and thus there is only a small area (0.1 mm) of coagulation beyond the vaporization crater.

The Nd YAG beam can produce coagulation at a depth of up to 6 mm into tissue, but superficial vaporization will occur if sufficient energy is used. At lower powers only coagulation will occur. This is very much in contrast to the CO&sub2; laser: a photon of light from the Nd YAG laser is 10 times more likely to be scattered than absorbed, and thus it will travel further into the tissue before it is absorbed.

In certain circumstances the thermal effects of the laser beam are not required, but photons are required to drive photochemical reactions in tissue, similar to those which are the basis of photosynthesis in plants. Such reactions in tissue produce important biological effects at power and energy levels below those required to produce thermal damage. The most promising technique involves the administration of photosensitizing agents which are retained with some selectivity in malignant tissue. When activated by light of a wavelength that is absorbed mainly by the photosensitizer and less by non-specific tissue components, a cytotoxic substance (singlet oxygen) is produced and tissue destruction occurs. The higher concentration of photosensitizer in malignant tumours offers the possibility of selective tumour destruction.

Other non-thermal laser effects are the ‘non-linear’ reactions that occur when tissue is exposed to pulsed laser light. The excimer lasers (ultraviolet wavelength, 175–355 nm) produce a laser beam in which the individual photon energy is very high and is highly absorbed by most biological tissues. This combination means that the light beam is capable of breaking interatomic bonds, and chemical photoablation of tissue occurs. The important biological feature of photoablation is the very sharp cut-off (of the order of a few microns) between normal cells and ablated tissue: there is no charred zone, as occurs with thermal laser ablation. The Nd YAG laser can be made to emit very short laser pulses which, if focused, can produce very high energies in a very small area. These high energies strip electrons from atoms and produce a rapidly expanding plasma of ions. This expanding plasma can generate powerful mechanical forces that can disrupt tissues. A ‘non-linear’ effect is used for laser lithotripsy in the endoscopic fragmentation of biliary calculi. Pulsed Nd YAG and dye laser beams are transmitted down flexible optical fibres, the ends of which are placed just touching the stone. The laser pulses produce a localized shock wave that can pulverize stones.

Upper gastrointestinal tract haemorrhage is responsible for up to 100 acute hospital admissions per 100000 of the population per annum in the United Kingdom. Over 60 per cent of these are due to peptic ulcer haemorrhage. Although most stop spontaneously there is still significant mortality and morbidity associated with continued or repeated bleeding, and to arrest haemorrhage in frail patients is also a source of morbidity. Surgery may be avoided if the laser is used as a endoscopic haemostatic device to produce thermal contraction of the bleeding vessel.

The bleeding or visible vessel is identified, any clot washed off, and the lesion is ringed with several pulses of laser energy (50–80 W in 1-s pulses). The tissue turns white when it is coagulated and the feeding vessel has been occluded. Controlled trials have confirmed the efficacy of this treatment, with 1 per cent mortality in a laser-treated group compared with 12 per cent in control patients. It is now clear that the Nd YAG laser is the most appropriate laser to use. It is likely that cheaper methods, particularly endoscopic injection around the bleeding vessel, will ultimately prove to be as effective.

Endoscopic laser therapy has also proved useful for the treatment of angiodysplasias of the upper gastrointestinal tract and colon. The major problem is identifying the lesion that has bled, since multiple abnormalities may be present. The principles of treatment are similar to that for bleeding peptic ulcers. First a circumferential ring of tissue around the lesion is treated to produce thermal contraction of any feeding vessels. Finally the lesion itself is coagulated.

Lasers can be used to arrest repeated bleeding from inoperable or recurrent gastric carcinoma but conventional laser therapy can be difficult because of the large surface requiring treatment. Recently it has been possible to treat some of the lesions with interstitial laser therapy, slowly coagulating the tumour by inserting a laser fibre at several points into its centre and using low power (1–5 W) for 100 to 1000 s.

Laser therapy is now widely used for the palliation of malignant dysphagia caused by oesophagogastric cancer and for the palliation of the symptoms produced by advanced, inoperable rectal cancer. Treatment is entirely local and the laser is used at high power (50–80 W in 1-s pulses) to coagulate and vaporize the tumour. Treatment starting at the distal extent of the tumour and working proximally (after dilatation if necessary) is safer than treatment starting at the upper end of the tumour, when the direction of the occluded segment is unclear. In the oesophagus there is a 5 per cent perforation rate but most of these can be managed conservatively (using intravenous fluids and antibiotics) without surgical intervention. The complication rate is lower than that of endoscopic intubation. This method is most suitable for the treatment of totally obstructing tumours and for tumours high in the oesophagus which are unsuitable for endoscopic prosthetic intubation. It is not useful if the tumour is extrinsic to the lumen or if a tracheo-oesophageal fistula is present: in these circumstances intubation is the preferred treatment. A further problem with laser therapy is that it must be repeated at monthly intervals in order to maintain the oesophageal lumen. The laser is also useful if there is obstruction by recurrent tumour growth following prosthetic tube insertion (Fig. 3) 1418.

Five per cent of patients with colorectal cancers have advanced metastatic spread or severe concomitant disease that render them unsuitable for surgery. Palliative colostomy will only bypass the obstruction and does little to relieve the local problems of discharge, bleeding, tenesmus, and incontinence. Local fulguration, cryotherapy, and transanal resection have provided some relief for these patients, but fulguration requires administration of a general anaesthetic and all techniques (except laser) are restricted to the management of low rectal cancers. Laser therapy aims to remove all of the exophytic tumour (Fig. 4) 1419, improving bleeding, discharge, and obstructive symptoms; incontinence is little improved and invasive pain is not helped at all. Occasionally laser therapy has been followed by prolonged survival. Measurement of quality of life in patients who have received palliative laser therapy has shown that little improvement occurs if the patient has pain as the predominant symptom or has only a short time to live (<10 weeks).

Colonoscopic Nd YAG laser therapy is also possible for benign colonic tumours. Although most polyps can be very effectively treated by snare diathermy, some large villous adenomas may be best treated with laser ablation. A large study reported complete eradication of 42 of 56 villous adenomas with the laser in patients monitored for up to 24 months.

Recently there has been an increasing interest in the use of interstitial laser hyperthermia for the treatment of solid tumours that are not resectable. Interstitial hyperthermia involves reducing the Nd YAG laser power from 50 to 80 W for 0.5 to 1 s to between 1 and 2 W delivered over a longer time (up to 1000 s) and inserting the fibre directly into tissue. Low power is used to avoid vaporization but produce hyperthermic destruction of the tissue. This technique has been investigated for the possible treatment of unresectable liver and pancreatic cancers. Multiple fibres are inserted into the tumour under ultrasound control. The progression of the damage can be monitored in real time by ultrasound or magnetic resonance, and the area of destruction can be matched to the size of the lesion.

Photodynamic therapy is an interesting new technique with the potential for selective destruction of cancers, based on the systemic administration of certain photosensitizing agents that are retained with some selectivity in malignant tissue. When exposed to laser light of an appropriate wavelength, a cytotoxic reaction occurs, causing cellular destruction. The retention of these agents appears to be related to non-specific tumour factors rather than to the photosensitizer used. In extracranial tissues the maximum tumour:normal tissue ratio that can be obtained with a variety of photosensitizing agents is between 2 and 3:1. Investigation of photodynamic therapy in experimental colorectal neoplasms has demonstrated an important biological advantage over thermal laser destruction: full thickness colonic damage produced by photodynamic therapy, unlike thermal damage, does not reduce the mechanical strength of the bowel or cause perforation, because the submucosal collagen is preserved. However selective necrosis is limited to a small area. Initial clinical application in the gastrointestinal tract has demonstrated its potential.

Lasers may be used for the fragmentation of biliary calculi that are too large to be removed by endoscopic sphincterotomy. The biliary tree is approached either by percutaneous transhepatic cholangioscopy or retrogradely following endoscopic sphincterotomy. Treatment is performed under sedation with fluoroscopic control with a laser fibre passed up the duct. Treatment can also be performed under direct vision using a small endoscope (the laser fibre passed through the instrumentation channel) passed through the instrumentation channel of a large duodenoscope into the common bile duct. Successful fragmentation is possible with subsequent passage of the fragments from the bile duct. It is conceivable that gallstones could be similarly treated without the need for sphincterotomy. There is the potential for laser treatment in recanalizing malignant biliary strictures which are impassable by a guidewire to allow placement of an endoprosthesis.

Barr H, Krasner N, Boulos PB, Chatlani P, Bown SG. Photodynamic therapy for colorectal cancer: a quantitative pilot study, Br J Surg, 1990; 77: 93–6.

Barr H, Krasner N. Interstitial laser photocoagulation for treating bleeding gastric cancer. Br Med J, 1990; 299: 659–60.

Bown SG, et al. Endoscopic treatment of inoperable colorectal cancers with the Nd YAG laser. Br J Surg, 1986; 73: 949–52.

Brunetaud JM, et al. Villous adenoma of the rectum. Results of endoscopic treatment with argon and Nd YAG lasers. Gastroenterology, 1985; 89: 832–7.

Krasner N, Barr H, Skidmore C, Morris AI. Palliative laser therapy for malignant dysphagia. Gut, 1987; 28: 792–8.

Lux G, et al. The first endoscopic retrograde laser lithotripsy of common bile duct stones in man using a pulsed Nd YAG laser. Endoscopy, 1986; 18: 144.

Rutgeerts P, Broeckaert L, Janssens J, Vantrappen G, Coremans Hiele M. Comparison of endoscopic polidocanol injection and YAG laser therapy for bleeding peptic ulcers. Lancet, 1989; ii: 1164–6.

Steger AC, Lees WR, Walmsley K, Bown SG. Interstitial laser hyperthermia: a new approach to local destruction of tumours. Br Med J, 1989; 299: 362–5.

Swain CP, Kirkham JS, Salmon PR, Bown SG, Northfield TC. Controlled trial of Nd YAG laser photocoagulation in bleeding peptic ulcers. Lancet, 1986; i: 1113–7.