Rumsfeld revisited: knowns and unknowns affecting the right heart

…because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don't know we don't know. – Donald H. Rumsfeld, US Secretary of Defense, 2002 (www.defense.gov/Transcripts/Transcript.aspx?TranscriptID=2636 (accessed 10/10/14) In this edition of Anaesthesia, Pilkington et al. present a timely review of pulmonary hypertension and its management in patients undergoing non-cardiac surgery 1. The review provides information and guidance for the management of patients who have a documented diagnosis of pulmonary hypertension where the specific cause of the condition and associated cardiac function, especially right ventricular function, has been previously assessed. They highlight the increased peri-operative morbidity and mortality in this group of patients and stress the integral role of the right ventricle, and its matching with the pulmonary circulation, in the prevention of such complications. It is striking that many of the associated risk factors identified in the review are also commonly seen in patients presenting for emergency surgery, particularly, surgery for hip fracture and emergency laparotomy. The question arises as to what influence undiagnosed pulmonary hypertension, and in particular, right heart dysfunction per se, may have on the outcome of these procedures. In the UK, the prevalence of heart failure is 0.9% in men and 0.7% in women, with a sharp rise with increasing age: 13.1% in men and 11.9% in women aged ≥ 75 years 2. The most common cause of heart failure in the UK is coronary artery disease which, in turn, has a prevalence in the UK population of 3.5% 3. The specific prevalence of right heart failure is much more difficult to identify. The National Pulmonary Hypertension Audit 4 suggests a prevalence of 124 patients per million people in the UK, but this is likely to be a significant underestimate, given that prevalence is based upon holding an active referral to one of the nationally designated pulmonary hypertension centres during the audit period. Again, the distribution of patients is skewed towards the elderly, peaking in the seventh and eighth decades. The right ventricle tends to respond better to volume overload (i.e. increased preload) than to increases in afterload. The consequences of an acute worsening in right ventricular function are profound, and may be life-threatening. A failing right ventricle may lead to arrhythmias, left ventricular dysfunction (secondary to interdependence), low cardiac output, hypotension and shock. Hence, the right ventricle should not be dismissed as an innocent bystander in cardiac dysfunction. Hip fracture has been identified as the commonest cause for surgery and anaesthesia in the frail elderly population. In 2013, the National Hip Fracture Database reported that approximately 70% of patients were ≥ 80 years old, with a 30-day hospital mortality of 8.2% 5. The 2014 Anaesthesia Sprint Audit of Practice collated data on co-morbidities in this patent population 6. Heart failure and pulmonary hypertension were not addressed individually; however, conditions associated with coronary artery disease and heart failure (namely angina, myocardial infarction, atrial fibrillation, valvular heart disease and hypertension) were widespread, as were conditions associated with pulmonary hypertension (aforementioned conditions and respiratory disease). The audit demonstrated a high prevalence of intra-operative hypotension, and concluded that 19% of patients receiving cemented hemi-arthroplasties may have experienced some level of bone cement implantation syndrome (BCIS). Reactions involving severe hypoxia and/or hypotension, or cardiovascular collapse, were documented in 2.7% and 0.5% of operations, respectively 7. Recognised risk factors for the development of BCIS include pre-existing pulmonary hypertension and significant cardiac disease 8. The proposed pathology of BCIS is via an increase in pulmonary vascular resistance, with subsequent impairment of right ventricular function and distension, thereby impairing left ventricular function and cardiac output by altering ventricular interdependence (the concept that the size, shape and function of one ventricle impacts upon the other). The adverse effect on ventricular function would be worsened in the presence of pre-existing right heart impairment and/or pulmonary hypertension. The resultant implication is that elderly patients who present for hip fracture surgery may have undiagnosed right heart dysfunction/pulmonary hypertension, and thus be at increased risk of peri-operative morbidity and mortality, and potentially at increased risk of conditions such as BCIS 9. The review by Pilkington et al. highlights the increased morbidity and mortality in those patients with diagnosed pulmonary hypertension presenting for non-cardiac surgery, and emphasises the pivotal role of the right ventricle and its management when it begins to fail 1. Thus, it is imperative that we raise awareness, increasing the index of suspicion for undiagnosed right heart dysfunction and pulmonary hypertension when we assess patients for emergency surgery, especially, one could argue, for patients who require a cemented prosthesis for hip fracture surgery. National audit has also afforded greater knowledge of morbidity and mortality associated with emergency laparotomy: the Emergency Laparotomy Network has reported an unadjusted 30-day mortality of 14.9% for all patients, increased further in patients with higher (worse) ASA physical status and in the elderly 10. However, identifying the pathological process(es) leading to individual deaths is less clear. Postmortem data are not routinely available or reported; it is therefore not known how many patients die in whom an undiagnosed clinical condition like pulmonary hypertension and right heart failure may have played a major role. Identifying right heart dysfunction, or a right ventricle that is ‘at risk’, depends upon the index of suspicion. The presence of a condition linked with pulmonary hypertension, as described by Pilkington et al., should raise the suspicion of both the presence of pulmonary hypertension and the co-existence of right heart dysfunction 1. Ischaemic heart disease or prior myocardial infarction, especially with ischaemia/infarction of the right ventricle, or lesions in those vessels supplying the right side (right coronary artery or left circumflex artery in a left-dominant circulation), should heighten our level of awareness. Elderly patients with multiple co-morbidities can be difficult to assess, since different disease processes share common symptoms. Breathlessness is the main symptom of pulmonary hypertension but can be due to a variety of important clinical conditions, as can syncope and chest pain. Reduced exercise tolerance, fatigue, oedema and palpitations are not specific to right heart dysfunction, but should heighten suspicion in the elderly patient presenting for emergency surgery. Right ventricular failure is a spectrum of disease, progressing from those patients deemed at risk, to asymptomatic right heart dysfunction, to symptomatic and thereafter refractory cases 11. In severe disease, signs of decompensated right heart failure may be evident, including tachycardia, arrhythmias, elevated jugular venous pressure, tricuspid regurgitation, hepatomegaly, ascites and ankle oedema. Investigations include electrocardiography, which may show signs of right heart strain. The chest X-ray may be normal, unless there is advanced disease. Right heart/pulmonary artery catheterisation may be of value in the diagnosis of pulmonary hypertension, but would prove impractical in assessing patients for hip fracture surgery and emergency laparotomy. Diagnostically, transthoracic echocardiography (TTE) is used extensively to assess cardiac function and to estimate pulmonary artery pressures 12. Its effectiveness is hampered by: image acquisition; the load dependence of right ventricular function, making assessment of the right ventricular function challenging 13; and the need for a trained individual to acquire and assess TTE images in a timely manner. Quoted reasonable indices of right ventricle contractility include right ventricle ejection fraction, right ventricle fractional area change, tissue Doppler strain analysis, and tricuspid annular plane systolic excursion (TAPSE), with the latter increasingly performed. B-Type natriuretic peptide (BNP) estimation is a relatively simple biochemical marker test used in the diagnosis and prognosis of left ventricular dysfunction. In addition, BNP appears to be an independent predictor for postoperative cardiac mortality in non-cardiac surgery 14. A study in patients with acute pulmonary embolism demonstrated the usefulness of BNP in identifying right ventricular dysfunction in normotensive patients with pulmonary embolism and no left ventricular dysfunction 15. The authors concluded that elevated BNP provided a useful marker of right ventricular overload in pulmonary embolism, thereby directing these patients for urgent TTE. In patients with pulmonary arterial hypertension, the biomarker high-sensitive cardiac troponin T (hsTnT) has been demonstrated to be associated with advanced World Health Organization functional class, impaired six-minute walk test and mortality, and is related to systolic right ventricular dysfunction 16. Future research may address whether BNP and hsTnT have a role in identifying patients with pulmonary hypertension and/or right heart dysfunction pre-operatively. If we suspect right ventricular dysfunction and/or pulmonary hypertension either by means of history, examination or previous investigation, how should we proceed? In undiagnosed patients presenting for emergency surgery, pragmatism must prevail. Patients with hip fracture have been demonstrated to have reduced morbidity and mortality when they are operated upon in a timely manner; it is important not to delay with investigations whose results will not change management. Transthoracic echocardiography is non-invasive and gives salient and valuable information; in addition, improvements in technology now mean that the TTE machine itself is cheaper and much more portable, with studies readily undertaken on the ward or pre-admission clinic. Learning the technique is now more accessible and practical for anaesthetists, and this is especially important considering their potential role as peri-operative physicians. If there is a valid suspicion of right heart dysfunction, this modality is beneficial in assessing both severity and cause. Much has been debated upon the value of TTE in hip fracture patients. Recent publications lean towards the benefit of using the modality in this group of patients 17-19, but reticence remains within the anaesthetic community. One could claim that assessment of the right ventricle gives fortitude to the argument for using TTE and strengthens our resolve to find appropriate ways to enable us to do so, including considering training and accreditation in echocardiography 20. The pragmatic approach dictates that if right ventricular dysfunction is suspected but not confirmed, then one must proceed with a pertinent knowledge of potential issues and an awareness of appropriate intra-operative management goals. Selecting appropriate monitoring is important: invasive arterial monitoring of blood pressure is, in our opinion, mandatory in patients with known or suspected right ventricular dysfunction and/or pulmonary hypertension. Echocardiography should also be considered, as it may provide valuable information on filling status, myocardial contractility and estimated pulmonary artery pressures, especially in the scenario where hypotension is developing. Transoesophageal echocardiography (TOE) is ideal in patients who are receiving mechanical ventilation, but intermittent TTE examination is appropriate in spontaneously breathing patients or when TOE skills or equipment are not available. Although limited as a diagnostic modality in this patient group, pulmonary artery catheterisation provides another monitoring option. Controversy surrounding its general use has led to a decline in practice over the years; however, judicious use of the pulmonary artery catheter in patients with confirmed severe pulmonary hypertension, or where there is severe peri-operative cardiovascular instability, can give appropriate and timely information on pulmonary artery pressures and the effect of therapeutic interventions. Hypotension is not well tolerated by the patient with right ventricular dysfunction and/or pulmonary hypertension, and may cause myocardial ischaemia and a downwards spiral of further hypotension and cardiovascular collapse. This is especially pertinent at the time of cemented prosthesis insertion, with the risk of BCIS. Meticulous attention should be paid to preload, afterload, myocardial contractility and the maintenance of sinus rhythm; bradycardia should also be avoided as the right ventricle may get over-stretched, and a heart rate between 70 and 100 beats per minute is desirable. Factors determining pulmonary vascular resistance must be optimised. Pharmacological management includes: the use of systemic and/or inhaled pulmonary artery vasodilators, such as iloprost (prostacyclin), to decrease right ventricle afterload; vasoconstrictors to optimise systemic arterial pressure and coronary perfusion; and optimal fluid loading and inotropes to improve both right and left ventricular contractility. The choice of pharmacological agents should be guided by monitoring, and the priority is maintaining adequate perfusion pressure; we suggest targeting a mean arterial pressure of at least 70 mmHg in patients with known right ventricular dysfunction. Postoperatively, critical care facilities should be available. Factors influencing pulmonary vascular resistance and thus right ventricular afterload should also be optimised in the postoperative environment, as they were in the operating theatre. Fortunately, the right ventricle appears to be relatively resilient to irreversible insult; following acute right ventricle myocardial infarction, right ventricular ejection fraction appears to increase significantly in the recovery phase, with the associated implication that myocardial stunning plays a pivotal role 21. Hence, timely intervention to support both the right and left ventricles, theoretically at least, can lead to recovery of right ventricular function following significant insult, providing further stimulus to consider right ventricular dysfunction/pulmonary hypertension in the peri-operative period. Without such intervention, right ventricular function may deteriorate to an irreversible extent in the peri-operative period, with ensuing end-organ failure and ultimately death as a result of cardiac failure and hypoperfusion. An appreciation of the importance of the right ventricle, coupled with a high index of suspicion when assessing elderly patients presenting for emergency surgery, especially if their history or clinical examination suggests impaired exercise tolerance or congestive cardiac failure, and employing appropriate supportive management of right ventricular dysfunction, may toget-her provide a further avenue for promoting safe emergency surgery in the more vulnerable patient populations. We should aim to move right heart function from an ‘unknown unknown’ to a ‘known unknown’ and, ultimately, a ‘known known’. No external funding and no competing interests declared.