Pulmonary hypertension due to chronic hypoxia Treat the lung not the pressure Pulmonary vasoconstriction in response to hypoxia is a basic reflex that occurs in all mammals studied. It has long been thought to have two main physiological functions. In the fetus it is responsible for closing off the pulmonary circulation to prevent blood from shunting through the unventilated lung. After birth it ensures that perfusion is constantly matched to ventilation. The problem comes when the lung is faced with global hypoxia due either to high altitude or to chronic hypoxic lung disease. Then the reflex seems malign and causes pulmonary hypertension. Not surprisingly, hypoxic vasconstriction is virtually absent in those animals that live constantly at high altitude. Bar headed geese, for example, fly over the Himalayas to nest in Tibet. These birds and also mammals that live at high altitude seem to suffer very little from their inability to control the flow of blood to the underventilated lung-so is the reflex of any benefit to adult animals or man? Patients with chronic hypoxic lung disease (usually chronic bronchitis and emphysema but also chronic asthma and restrictive diseases such as fibrosing alveolitis and kyphoscoliosis) are all at risk of pulmonary hypertension, and its development is associated with a poor prognosis.' Some authors have claimed that the pulmonary hypertension in these patients is coincidental to their poor prognosis and poor exercise tolerance rather than their cause. They base this assertion on the observation that the resting pulmonary artery pressure is often not high in patients with severe lung disease. A mean pressure of only 30 mm Hg is typical-whereas in primary pulmonary hypertension it may often reach systemic values. In such patients, however, the pulmonary pressure may rise much higher on exercise, owing to the low compliance of the constricted pulmonary circulation, or at night, owing to worsening hypoxia. Furthermore, the low oxygen content of the arterial blood means that adequate oxygen delivery to the tissues depends on a high cardiac output, and these patients are often unable to raise their cardiac outputs fuirther on exercise because of the increased afterload.2 So, if pulmonary hypertension in chronic hypoxic lung disease is important, how can we recognise it, and how should we treat it? Typically the patient with secondary pulmonary hypertension will complain of current or past episodes of fluid retention and will have fixed airflow obstruction (forced expiratory volume in one second less than 1 5 litres), hypoxaemia (arterial pressure of oxygen below 8 kPa), and hypercapnia (arterial carbon dioxide pressure over 6 kPa). The chest radiograph may show an enlarged pulmonary artery and the electrocardiogram may show P pulmonale, right axis deviation, and dominant R waves in lead VI. In the end the diagnosis depends on the findings at right heart catheterisation, and this should be available in any specialist respiratory unit. In the future new techniques of Doppler echocardiography should provide the same information non-invasively, but at present adequate Doppler signals are seen in only about a third of patients because of interference by the overlying lung (unpublished data). The treatment of secondary pulmonary hypertension has centred on the use of vasodilators including [32 agonists, hydralazine, isosorbide, calcium antagonists, and angiotensin converting enzyme (ACE) inhibitors.`- Of the vasodilators the ACE inhibitors might be expected to be the most effective BMJ
VOLUME 300
24 MARCH 1990
because they have the potential to improve two aspects of the disorder-namely, the pulmonary hypertension and the fluid retention, which is thought to be due to the effects of hypoxaemia and hypercapnia on the kidney.'0 The problem with all the vasodilators (other than oxygen) is that they are non-selective and cause unwanted systemic vasodilatation, which limits their use. There are hopes that selective vasodilators will be developed; until they are, no further studies of vasodilators need be done. Their limited value implies that, at present, treatment should concentrate on maximising lung function and improving oxygenation, if necessary by the use of domiciliary oxygen. All patients with chronic hypoxic lung disease and an episode of fluid retention should be referred to a specialist unit. There they should have lung function studies including assessment of the extent of the reversal of the airflow obstruction by f agonists, anticholinergics, and steroids and measurement of arterial blood gas tensions. If appropriate these patients should also be assessed for suitability for domiciliary treatment with oxygen, should undergo measurement of pulmonary haemodynamics, and have sleep studies to rule out nocturnal deterioration of hypoxaemia. Remarkably, often patients' pulmonary haemodynamics and exercise tolerance will be improved by these measures. ANDREW PEACOCK
Senior Registrar in Respiratory Medicine, Southampton General Hospital, Southampton S09 4XY I Weitzenblum E, Hirth C, Ducolone A, Mirhom R, Rasaholinjanahary J, Ehrhart MN1. Progniostic value of pulmonary artery pressure in chronic obstructive lung disease. Thorax 1981;36:752-8. 2 Morrison DA. Pulmonary hypertension in chronic obstructive pulmonary disease: the right ventricular hypothesis. Chest 1987;92:387-9. 3 Danahv DT, Tobis JM, Aronow WS, Chetty K, Glauser F. Eff'ects of isosorbide dinitrate on pulmonary hypertension in chronic obstructive pulmonary disease. Cnli Pharmacol I'her
1979;25:541-8. 4 Peacock AJ, Busst C, Dawkins K, Denison DM. Response of the pulmonary circulation to oral pirbuterol in chronic airflow obstruction. BrMedJ7 1983;287:1178-80. 5 Macnee W, Wathen CG, Hannan WJ, Flenley DC, Muir Al.. Effects of' pirbuterol and sodium nitroprusside on pulmonary haemodynamics in hvpoxic cor pulmonale. Br Mfed .7 1983;287: 1169-72. 6 Keller CA, Shepard JW Jr, Chun t)S, Dolan GF, Vasquez P, Mitnh V-D. Effects of hvdralazitsc tttt haemodynamics, ventilation, and gas exchange in patietats with chronic obstructive ptltmonary disease and pulmonary hypertension. A.m Rev Respir Dis 1984;130:606-11. 7 Kennedy TPl, Michael JR, Chun-Kang H. Nifedipine inhibits hypoxic pulmonary sasoconstriction during rest and exercise in patients with chronic obstructive pulmonary disease. Am Rev
Respirl)is 1984;129:544-51. 8 Mu1tramoto A, Caldwell J, Albert RK, Lakshminarayatn S, 13utler J. Nifedipine dilates thy ptlmonary vasculature without producing symptomatic systetnic hypotension in upright resting and exercising patients with pulmonary hypertensiotn seconidary to chronic obstructiVe pulmonary disease. Am Rez' Respir Dis 1985;132:963-6. 9 Peacock A, MNatthews A. ''he effect of captopril on pulmonary hacmodvnamics and lung function ilt patients with chronic airflow obstruction. Thorax 1986;41:225. 10 Olixer RM, Peacock AJ, Fleming JS, Waller D. The renal anid pulmonary effects of captopril in patients with hypoxic lung disease. T'horax 1989;44:513-5.
Varicose veins An increasing burden for the NHS Varicose veins are a major and increasing burden to the health service. Over 50000 patients were admitted to hospital for treatment in 1987-8'-almost twice as many as in 1975.2 Many hospitals have huge waiting lists for operations on veins, and some surgeons will not see new patients with this complaint. A low level of interest by doctors has been suggested as the cause of this state of affairs.' The primary causes, however, are the high prevalence of varicose veins4 5 and the substantial time required for treatment of a condition that is not a serious threat to health. The term varicose veins includes anything from minor cosmetic blemishes to important venous hypertension with 763
ulcers. Most patients simply have unsightly veins, heaviness and aching of the legs, or mild ankle swelling; they need to be reassured that their condition is medically harmless, and some will then decide against active treatment.4 Some opt for support stockings, which are effective in relieving heaviness and aching.6 Most, however, find conservative management unacceptable; they want to be rid of their veins. Careful assessment is imporant for patients who want treatment. Ordinary clinical examination may be adequate, but most vascular surgeons now use a Doppler ultrasound probe to listen to the flow in the veins during manoeuvres such as calf squeezing, the Trendelenberg test, and removal of venous tourniquets.7'0 The Doppler probe is particularly helpful in examining the short saphenous system" and recurrent varicose veins.7 The aim is to detect incompetence of values between the deep venous- system and the varicose veins. Incompetent valves occur most commonly at the saphenofemoral junction in the groin, but they also occur in the short saphenous vein behind the knee and in the direct perforating veins in the calf or thigh. If these are not identified and dealt with the varicosities will recur.7 1213 Other methods of assessing the veins'- such as venography or venous function tests -are not usually necessary for simple varicose veins but may be needed when there is a history ofdeep vein thrombosis, calf perforator incompetence, or surgery.7 12 If there is no upstream incompetence of valves then injection treatment (compression sclerotherapy) is effective, especially for varicose veins below the knee, where good compression bandaging can be applied. " Sclerotherapy clinics offer good early results to large numbers of outpatients at the expense of a few weeks of bandaging.'5'8 If the varicosities recur injections can be repeated. Recurrence is common, however, because injections are often given to patients with incompetent valves, and over half of them will suffer recurrent varicose veins within five years. I When there are incompetent valves between the deep and superficial veins surgery provides the best long term treatment.7 117 These sites of incompetence are dealt with by dividing the main superficial vein flush with the deep veins and also dividing any nearby tributaries that might offer a route for recurrence.'3 The long saphenous vein is usually stripped from groin to upper calf,9 1920 especially if it is joined by an incompetent perforating vein in the lower thigh.'22' Stripping to the ankle is unnecessary22 and often damages the saphenous nerve, causing some numbness in the foot.'9 All varicosities should be carefully marked before operation and may then be removed by avulsing them through tiny incisions. Sutureless closure avoids stitch marks and provides the best cosmetic result.20 Bilateral varicose vein operations are time consuming, and patients who need this procedure contribute substantially to long waiting lists. A team approach to surgery helps to save time; these are not operations for a solo surgical trainee. Unilateral procedures are certainly suitable for day case surgery, and some surgeons will operate on both legs in'day patients.2324 A few perform simple saphenofemoral ligation, which is followed later by outpatient sclerotherapy; but this is a more complicated approach for the patient.23 Even after bilateral surgery patients should be mobilised and return to full normal activity as early as possible. The traditional advice
764
simply to "walk three miles a day" needs modification to encourage frequent activity rather than a single long walk. Clear priorities are essential in managing the many patients presenting with varicose veins. Those with skin changes, ulcers, recurrent phlebitis, or a history of bleeding should be seen without undue delay, fully assessed, and treated. The many more with cosmetic worries or discomfort cannot be given priority over more urgent problems, and their management will depend on local -circumstances. Ideally, they should be expertly assessed, counselled on the advantages and disadvantages of treatment, and offered whatever treatment they wish. In practice the choice is often between temporary improvement by sclerotherapy or a long wait for surgery. Such patients should be referred to hospital only after reassurance that their varicose veins are harmless and if they are still keen to have active treatment. W BRUCE CAMPBELL
Consultant in Vascular and General Surgery, Royal Devon and Exeter Hospital, Exeter EX2 5DW I Department of Health and Social Secutrity. Hospital episode statistics 1987-88. London: DHSS, 1988. 2 Department of Health and Social Security and Office of Population Censuses and Surveys. Hospital In-patient Enquirv 1975. London: HMSO, 1983. (Series MB4 (15).) 3 Frankel S. The natural history of waiting lists-some wider explanations for an unnecessary problem. Health Trends 1989;21:56-8. 4 Madar G, Widmer LK, Zemp E, Maggs M. Varicose veins and chronic venous insufficiency: disorder or disease? Vasa 1986;15: 126-34. 5 Brand FN, Dannenberg AL, Abbott RD, Kannel WB. The epidemiology of varicose veins: the Framingham study. AmJ Prev Med 1988;4:%-101. 6 Chant ADB, Magnussen P, Kershaw C. Support hose and varicose veins. BrMedJ 1985;290:204. 7 Royle JP. Recurrent varicose veins. World Surg 1986;10:944-53. X Bladin C, Royle JP. Acquisition of skills required for use of Doppler ultrasound and the assessment of varicose veins Aust N ZJ7 Surg 1987;57:225-6. 9 Eklof B. Modern treatment of varicose veins. BrJ Surg 1988;75:297-8. 10 Ruckley CV. A colour atlas of surgical management of venous disease. London: Wolfe Medical Publicatioins, 1988. 11 MAitchell [)C, Darke SG. [ he assessment of primary varicose veins by Doppler ultrasound-the role of saphenopopliteal incompetence and the short saphenous systems in calf varicosities. Eurj Vasc Surg 1987;1: 113-5. 12 Corbett CR, Mclrvine AJ, Aston NO, Jamieson C(W. Lea-Thomas.M. The use of varicography to identify the sources of incompetence in recurrent varicose veins. Attn R Coll Surg Engl
1984;66:412-5. 13 Starnes HF, Valiance R, Hamilton DNH. Recurrent varicose veins: a radiological approach to investigation. Clin Radial 1984;35:95-9. 14 Fegan WG. Continuous compression technique of injecting varicose veins. Lancet 1963;ii: 10)9-12. 15 Reid RG, Rothnie NG. Treatment of varicose veins by compression sclcrotherapy. Br 7 Surg 1968;55:889-95. 16 Hobbs ST. Surgery and sclerotherapy in the treatment of varicose veins: a random trial. Arch Surg 1974;109:793-6. 17 Jakobsen BH. The salue of different forms of treatment for varicose veins. Br J Surg 1979,66: 182-4. 18 Batch AJG, Wickremesinghe SS, Gannon ME, Dormandy JA. Randomised trial of bandaging after sclerotheNapy for varicose veins. BrMedj 1980;281:423. 19 Munn SR, Morton JB, Macbeth WAAG, McLeish AR. To strip or not to strip the long saphenous vein? A varicose veins trial. Brj Surg 1981;68:426-8. 20 Rivlin S. The surgical cure of primary varicose veins Brj Surg 1975;62:913-7. 21 Sutton R, Darke SG. Should the long saphenous vein be stripped? A study by per-operative retrograde saphenography. In: Negus D, Jantet G, eds. Phlebology '85. London: John Libbey, 1986:196-9. 22 Macfarlane R, Godwin RJ, Barabas AP. Are varicose veins and coronary artery bypass surgery compatible? Lancet 1985;ii:859. 23 Bishop CCR, Jarrett PEM. Outpatient varicose vein surgery under local anaesthesia. Brj Surg 1986;73:821-2. 24 Neglen P, Einarsson E, Jonsson B, Eklof B. Socioeconomic benefits of ambulatory surgery or compression sclerotherapy of varicose veins. In: Negus D, Jantet G, eds. Phlebology '85. London: John Libbey, 1986:159-62.
Correction Non-steroidal anti-inflammatory drugs and peptic ulcers An author's error occurred in this regular review by Dr C J Hawkey (3 February, p 283). In figure 6 the dosage of ranitidine should have been given as 150 mg twice daily, not 150 [tg as published.
BMJ voi umiF 300
24 MARCh' 1990
VOLUME 300
24 MARCH 1990
because they have the potential to improve two aspects of the disorder-namely, the pulmonary hypertension and the fluid retention, which is thought to be due to the effects of hypoxaemia and hypercapnia on the kidney.'0 The problem with all the vasodilators (other than oxygen) is that they are non-selective and cause unwanted systemic vasodilatation, which limits their use. There are hopes that selective vasodilators will be developed; until they are, no further studies of vasodilators need be done. Their limited value implies that, at present, treatment should concentrate on maximising lung function and improving oxygenation, if necessary by the use of domiciliary oxygen. All patients with chronic hypoxic lung disease and an episode of fluid retention should be referred to a specialist unit. There they should have lung function studies including assessment of the extent of the reversal of the airflow obstruction by f agonists, anticholinergics, and steroids and measurement of arterial blood gas tensions. If appropriate these patients should also be assessed for suitability for domiciliary treatment with oxygen, should undergo measurement of pulmonary haemodynamics, and have sleep studies to rule out nocturnal deterioration of hypoxaemia. Remarkably, often patients' pulmonary haemodynamics and exercise tolerance will be improved by these measures. ANDREW PEACOCK
Senior Registrar in Respiratory Medicine, Southampton General Hospital, Southampton S09 4XY I Weitzenblum E, Hirth C, Ducolone A, Mirhom R, Rasaholinjanahary J, Ehrhart MN1. Progniostic value of pulmonary artery pressure in chronic obstructive lung disease. Thorax 1981;36:752-8. 2 Morrison DA. Pulmonary hypertension in chronic obstructive pulmonary disease: the right ventricular hypothesis. Chest 1987;92:387-9. 3 Danahv DT, Tobis JM, Aronow WS, Chetty K, Glauser F. Eff'ects of isosorbide dinitrate on pulmonary hypertension in chronic obstructive pulmonary disease. Cnli Pharmacol I'her
1979;25:541-8. 4 Peacock AJ, Busst C, Dawkins K, Denison DM. Response of the pulmonary circulation to oral pirbuterol in chronic airflow obstruction. BrMedJ7 1983;287:1178-80. 5 Macnee W, Wathen CG, Hannan WJ, Flenley DC, Muir Al.. Effects of' pirbuterol and sodium nitroprusside on pulmonary haemodynamics in hvpoxic cor pulmonale. Br Mfed .7 1983;287: 1169-72. 6 Keller CA, Shepard JW Jr, Chun t)S, Dolan GF, Vasquez P, Mitnh V-D. Effects of hvdralazitsc tttt haemodynamics, ventilation, and gas exchange in patietats with chronic obstructive ptltmonary disease and pulmonary hypertension. A.m Rev Respir Dis 1984;130:606-11. 7 Kennedy TPl, Michael JR, Chun-Kang H. Nifedipine inhibits hypoxic pulmonary sasoconstriction during rest and exercise in patients with chronic obstructive pulmonary disease. Am Rev
Respirl)is 1984;129:544-51. 8 Mu1tramoto A, Caldwell J, Albert RK, Lakshminarayatn S, 13utler J. Nifedipine dilates thy ptlmonary vasculature without producing symptomatic systetnic hypotension in upright resting and exercising patients with pulmonary hypertensiotn seconidary to chronic obstructiVe pulmonary disease. Am Rez' Respir Dis 1985;132:963-6. 9 Peacock A, MNatthews A. ''he effect of captopril on pulmonary hacmodvnamics and lung function ilt patients with chronic airflow obstruction. Thorax 1986;41:225. 10 Olixer RM, Peacock AJ, Fleming JS, Waller D. The renal anid pulmonary effects of captopril in patients with hypoxic lung disease. T'horax 1989;44:513-5.
Varicose veins An increasing burden for the NHS Varicose veins are a major and increasing burden to the health service. Over 50000 patients were admitted to hospital for treatment in 1987-8'-almost twice as many as in 1975.2 Many hospitals have huge waiting lists for operations on veins, and some surgeons will not see new patients with this complaint. A low level of interest by doctors has been suggested as the cause of this state of affairs.' The primary causes, however, are the high prevalence of varicose veins4 5 and the substantial time required for treatment of a condition that is not a serious threat to health. The term varicose veins includes anything from minor cosmetic blemishes to important venous hypertension with 763
ulcers. Most patients simply have unsightly veins, heaviness and aching of the legs, or mild ankle swelling; they need to be reassured that their condition is medically harmless, and some will then decide against active treatment.4 Some opt for support stockings, which are effective in relieving heaviness and aching.6 Most, however, find conservative management unacceptable; they want to be rid of their veins. Careful assessment is imporant for patients who want treatment. Ordinary clinical examination may be adequate, but most vascular surgeons now use a Doppler ultrasound probe to listen to the flow in the veins during manoeuvres such as calf squeezing, the Trendelenberg test, and removal of venous tourniquets.7'0 The Doppler probe is particularly helpful in examining the short saphenous system" and recurrent varicose veins.7 The aim is to detect incompetence of values between the deep venous- system and the varicose veins. Incompetent valves occur most commonly at the saphenofemoral junction in the groin, but they also occur in the short saphenous vein behind the knee and in the direct perforating veins in the calf or thigh. If these are not identified and dealt with the varicosities will recur.7 1213 Other methods of assessing the veins'- such as venography or venous function tests -are not usually necessary for simple varicose veins but may be needed when there is a history ofdeep vein thrombosis, calf perforator incompetence, or surgery.7 12 If there is no upstream incompetence of valves then injection treatment (compression sclerotherapy) is effective, especially for varicose veins below the knee, where good compression bandaging can be applied. " Sclerotherapy clinics offer good early results to large numbers of outpatients at the expense of a few weeks of bandaging.'5'8 If the varicosities recur injections can be repeated. Recurrence is common, however, because injections are often given to patients with incompetent valves, and over half of them will suffer recurrent varicose veins within five years. I When there are incompetent valves between the deep and superficial veins surgery provides the best long term treatment.7 117 These sites of incompetence are dealt with by dividing the main superficial vein flush with the deep veins and also dividing any nearby tributaries that might offer a route for recurrence.'3 The long saphenous vein is usually stripped from groin to upper calf,9 1920 especially if it is joined by an incompetent perforating vein in the lower thigh.'22' Stripping to the ankle is unnecessary22 and often damages the saphenous nerve, causing some numbness in the foot.'9 All varicosities should be carefully marked before operation and may then be removed by avulsing them through tiny incisions. Sutureless closure avoids stitch marks and provides the best cosmetic result.20 Bilateral varicose vein operations are time consuming, and patients who need this procedure contribute substantially to long waiting lists. A team approach to surgery helps to save time; these are not operations for a solo surgical trainee. Unilateral procedures are certainly suitable for day case surgery, and some surgeons will operate on both legs in'day patients.2324 A few perform simple saphenofemoral ligation, which is followed later by outpatient sclerotherapy; but this is a more complicated approach for the patient.23 Even after bilateral surgery patients should be mobilised and return to full normal activity as early as possible. The traditional advice
764
simply to "walk three miles a day" needs modification to encourage frequent activity rather than a single long walk. Clear priorities are essential in managing the many patients presenting with varicose veins. Those with skin changes, ulcers, recurrent phlebitis, or a history of bleeding should be seen without undue delay, fully assessed, and treated. The many more with cosmetic worries or discomfort cannot be given priority over more urgent problems, and their management will depend on local -circumstances. Ideally, they should be expertly assessed, counselled on the advantages and disadvantages of treatment, and offered whatever treatment they wish. In practice the choice is often between temporary improvement by sclerotherapy or a long wait for surgery. Such patients should be referred to hospital only after reassurance that their varicose veins are harmless and if they are still keen to have active treatment. W BRUCE CAMPBELL
Consultant in Vascular and General Surgery, Royal Devon and Exeter Hospital, Exeter EX2 5DW I Department of Health and Social Secutrity. Hospital episode statistics 1987-88. London: DHSS, 1988. 2 Department of Health and Social Security and Office of Population Censuses and Surveys. Hospital In-patient Enquirv 1975. London: HMSO, 1983. (Series MB4 (15).) 3 Frankel S. The natural history of waiting lists-some wider explanations for an unnecessary problem. Health Trends 1989;21:56-8. 4 Madar G, Widmer LK, Zemp E, Maggs M. Varicose veins and chronic venous insufficiency: disorder or disease? Vasa 1986;15: 126-34. 5 Brand FN, Dannenberg AL, Abbott RD, Kannel WB. The epidemiology of varicose veins: the Framingham study. AmJ Prev Med 1988;4:%-101. 6 Chant ADB, Magnussen P, Kershaw C. Support hose and varicose veins. BrMedJ 1985;290:204. 7 Royle JP. Recurrent varicose veins. World Surg 1986;10:944-53. X Bladin C, Royle JP. Acquisition of skills required for use of Doppler ultrasound and the assessment of varicose veins Aust N ZJ7 Surg 1987;57:225-6. 9 Eklof B. Modern treatment of varicose veins. BrJ Surg 1988;75:297-8. 10 Ruckley CV. A colour atlas of surgical management of venous disease. London: Wolfe Medical Publicatioins, 1988. 11 MAitchell [)C, Darke SG. [ he assessment of primary varicose veins by Doppler ultrasound-the role of saphenopopliteal incompetence and the short saphenous systems in calf varicosities. Eurj Vasc Surg 1987;1: 113-5. 12 Corbett CR, Mclrvine AJ, Aston NO, Jamieson C(W. Lea-Thomas.M. The use of varicography to identify the sources of incompetence in recurrent varicose veins. Attn R Coll Surg Engl
1984;66:412-5. 13 Starnes HF, Valiance R, Hamilton DNH. Recurrent varicose veins: a radiological approach to investigation. Clin Radial 1984;35:95-9. 14 Fegan WG. Continuous compression technique of injecting varicose veins. Lancet 1963;ii: 10)9-12. 15 Reid RG, Rothnie NG. Treatment of varicose veins by compression sclcrotherapy. Br 7 Surg 1968;55:889-95. 16 Hobbs ST. Surgery and sclerotherapy in the treatment of varicose veins: a random trial. Arch Surg 1974;109:793-6. 17 Jakobsen BH. The salue of different forms of treatment for varicose veins. Br J Surg 1979,66: 182-4. 18 Batch AJG, Wickremesinghe SS, Gannon ME, Dormandy JA. Randomised trial of bandaging after sclerotheNapy for varicose veins. BrMedj 1980;281:423. 19 Munn SR, Morton JB, Macbeth WAAG, McLeish AR. To strip or not to strip the long saphenous vein? A varicose veins trial. Brj Surg 1981;68:426-8. 20 Rivlin S. The surgical cure of primary varicose veins Brj Surg 1975;62:913-7. 21 Sutton R, Darke SG. Should the long saphenous vein be stripped? A study by per-operative retrograde saphenography. In: Negus D, Jantet G, eds. Phlebology '85. London: John Libbey, 1986:196-9. 22 Macfarlane R, Godwin RJ, Barabas AP. Are varicose veins and coronary artery bypass surgery compatible? Lancet 1985;ii:859. 23 Bishop CCR, Jarrett PEM. Outpatient varicose vein surgery under local anaesthesia. Brj Surg 1986;73:821-2. 24 Neglen P, Einarsson E, Jonsson B, Eklof B. Socioeconomic benefits of ambulatory surgery or compression sclerotherapy of varicose veins. In: Negus D, Jantet G, eds. Phlebology '85. London: John Libbey, 1986:159-62.
Correction Non-steroidal anti-inflammatory drugs and peptic ulcers An author's error occurred in this regular review by Dr C J Hawkey (3 February, p 283). In figure 6 the dosage of ranitidine should have been given as 150 mg twice daily, not 150 [tg as published.
BMJ voi umiF 300
24 MARCh' 1990
