Oxford Handbook of Clinical Medicine (I)
“We wrote this book not because we know so much, but because we know we remember so little…the problem is not simply the quantity of information, but the diversity of places from which it is dispensed. Trailing eagerly behind the surgeon, the student is admonished never to forget alcohol withdrawal as a cause of post-operative confusion. The scrap of paper on which this is written spends a month in the pocket before being lost for ever in the laundry. At different times, and in inconvenient places, a number of other causes may be presented to the student. Not only are these causes and aphorisms never brought together, but when, as a surgical house officer, the former student faces a confused patient, none is to hand.”
‘But now you don’t need to look for those scraps of paper anymore because we’ve collected all that information right here, in this book,’ the authors would argue. Or at least some of the important information is included here (despite this being a 900+ page textbook, many books on subtopics covered in the book are much longer than that; for example the Holmes et al. textbook dealing only with sexually transmitted diseases is more than twice as long as this one. Of course a book with that kind of page count will only ever be a ‘handbook’ to someone with acromegaly…).
Anyway, I’m currently reading this book and I figured I should probably talk about a few of the observations made in the book here, to make them easier to remember later on. The book is intended to be used as a reference work for doctors so in a way trying to remember stuff written in it is a strange thing to do – the point of the book is after all that you don’t need to remember all that stuff – but I would prefer to remember some of the things written in this book and this’ll be easier to do if I write about them here on the blog, instead of just ‘keeping them hidden in the book’, so to speak.
I’m assuming nobody reading along here are planning on reading this book so I wasn’t sure how much sense it would make to add impressions about the way it’s written etc. here, but I decided to note down a few things on these topics anyway. I have noted along the way that the authors sometimes include comments about a condition which they only cover later in the same chapter, and this has bothered me a few times; on the other hand I’m well aware that when you’re trying to write a book where it’s supposed to be easy to look things up quickly you need to make some key decisions here and there which will be likely to impact the reading experience of people who read the book from cover to cover the way I am negatively. Most chapters are structured a bit the same way the ‘[Topic X] At a glance…’ textbooks I’ve read in the past were (Medical Statistics at a Glance, Nutrition at a Glance, The Endocrine System at a Glance); the chapters vary in length (for example there are roughly 70 pages about cardiovascular medicine, 40 pages about endocrinology, 50 pages about gastroenterology, and 30 pages about renal medicine) but they generally seem to be structured in much the same way; the chapters are segmented – many chapter segments are two-page segments, which were also predominant in the At a glance texts – and each segment deals with a specific topic in some detail, with details about many aspects of the disease/condition in question, such as information about e.g. incidence/prevalence, risk factors, some notes on pathophysiology, presentation/symptoms/signs, diagnostics (tests to perform, key symptoms to keep in mind, etc.), treatment options (drugs/surgery/etc.?, dosage, indications/contraindications, side effects, drug interactions, etc.), potential complications, and prognostic information. Not all chapters are structured in the ‘two-page-segments’ way even though this seems to be the baseline structure in many contexts; it’s clear that they’ve given some thought as to how best to present the information included in the coverage. I recall from the At a glance texts that I occasionally thought that the structure felt unnatural, and that they seemed to have committed to a suboptimal coverage format in the specific context – I have not thought along such lines while reading this book, which to me is a sign that they’ve handled these things well. Deviation from the default format occurs e.g. in the chapter on cardiovascular medicine, which has quite a few successive pages on which various types of ECG abnormalities are illustrated (I looked at that stuff and I like to think that I understand this stuff better than I used to now, but I must admit that this was one of the sections of this book into which I did not put a lot of effort, as it in some sense felt like ‘irrelevant knowledge’ – so don’t expect me to be able to tell a right bundle branch block from an acute anterior myocardial infarction on an EEG without having access to this book…). It’s perhaps important to point out that despite the condensed structure of the book the coverage is reasonably detailed; this is not a book with two pages about ‘heart disease’, it’s a book with two pages about rheumatic fever, two pages about right heart valve disease, two pages about infective endocarditis, two pages about broad complex tachycardia, etc. And many of the pages include a lot of information. I have read textbooks dealing with many of the topics they cover and this is also not my first general ‘clinical medicine’ text (that was McPhee et al.), but I’m learning new stuff from the book even about topics with which I’m familiar, which is really nice. It’s a pretty good book so far, even if it’s not perfect; I’m probably at a four star rating at the moment.
In the parts to follow I’ll talk about some of the observations included in the book which I figured might be worth repeating here.
The first observation: They note in the book that 80% of people above the age of 85 years (in Britain) live at home and that 70% of those people can manage stairs; they argue in the same context that any deterioration in an elderly patient should be considered to be from treatable disease until proven otherwise (i.e., the default should not be to say that ‘that’s probably just ageing’).
“Unintentional weight loss should always ring alarm bells”.
A diabetic is probably well-advised to be aware of some of the signs of peripheral arterial disease. These include loss of hair, pallor, shiny skin, cyanosis (bluish discoloration of the skin), dry skin, scaling, deformed toenails, and lowered skin temperature.
“Normally 400-1300mL of gas is expelled PR in 8-20 discrete (or indiscrete) episodes per day. […] most patients with ‘flatulence’ have no GI disease. Air swallowing (aerophagy) is the main cause of flatus; here N2 is the chief gas. If flatus is mostly methane, N2 and CO2, then fermentation by bowel bacteria is the cause, and reducing carbohydrate intake (eg less lactose and wheat) may help.”
If there are red blood cells in the urine, this is due to cancer or glomerulonephritis (let’s not go into details here – we’ll just call this one ‘kidney disease’ for now) until proven otherwise. Painless visual haematuria (blood in the urine) usually equals bladder cancer – it’s definitely a symptom one should have a talk with a doctor about. The book does not mention this, but it’s important to keep in mind however that red/brownish urine is not always due to blood in the urine; it can also be caused by drugs and vegetable dyes (link). I was very surprised about this one in the context of ways to prevent UTIs: “There is no evidence that post-coital voiding, or pre-voiding, or advice on wiping patterns in females is of benefit.” Drinking more water and drinking cranberry or lingo berry juice daily works/lowers risk.
Kidney function is often impaired in people who are hospitalized, with acute kidney injury (-AKI) occurring in up to 18% of hospital patients. It’s an important risk factor for mortality. Mortality can be very high in people with AKI, for example people admitted with burns who develop AKI have an 80% mortality rate, and with trauma/surgery it’s 60%. Up to 30 % of cases are preventable, and preventable causes include medications (continuing medications as usual e.g. after surgery can be catastrophic, and some of the drugs that can cause kidney problems are drugs people take regularly for chronic conditions such as high blood pressure or diabetes (metformin in particular)) and contrast material used in CT scans and procedures. Kidney function is incidentally often also (chronically) impaired in old people, most of which have no symptoms; “many elderly people fall into CKD [chronic kidney disease] stage 3 but have little or no progression over many years.” Symptoms of chronic kidney disease will usually not present until stage four is reached, but if onset of kidney failure is slow even people in the later stages may remain asymptomatic. The authors question whether it makes sense to label the old people in stage 3 with an illness; I’m not sure I completely agree (lowered kidney function increases cardiovascular risk, and some of those people may want to address this, if possible), but I’d certainly agree with the position that there’s a risk of overdiagnosis here.
A few more observations about kidneys. The chief cause of death from renal failure is cardiovascular disease, and in the first two stages of chronic kidney disease, the risk of dying from cardiovascular disease is higher than the risk of ever reaching stage 5, end-stage-renal-failure. Blood pressure control is very important in kidney disease as the authors argue that even a small drop in blood pressure may save significant kidney function. The causal link between BP and kidney disease goes both ways: “Hypertension often causes renal problems […] and most renal diseases can cause hypertension”. Once people require renal replacement therapy (RRT) such as haemodialysis mortality is high: Annual mortality is ~20%, mainly due to cardiovascular disease. The authors talk a little bit about diabetes and kidney disease in the book and among other things include the following observations:
“Diabetes is best viewed as a vascular disease with the kidney as one of its chief targets for end-organ damage. The single most important intervention in the long-term care of DM is the control of BP, to protect the heart, the brain, and the kidney. Renal damage may be preventable with good BP and glycaemic control.
In type 1 DM nephropathy is rare in the first 5yrs, after 10yrs annual incidence rises to a peak at 15yrs, then falls again. Those who have not developed nephropathy at 35yrs are unlikely to do so. In type 2 DM around 10% have nephropathy at diagnosis and up to half will go on to develop it over the next 20yrs. 20% of people with type 2 DM will develop ESRF.”
I was surprised by the observation above that “Those who have not developed nephropathy at 35yrs are unlikely to do so”, and I’m not sure I’d agree with the authors about that. The incidence of diabetes-related nephropathy peaks after a diabetes duration of 10-20 years and declines thereafter, but it doesn’t go to zero: “The risk for the development of diabetic nephropathy is low in a normoalbuminuric patient with diabetes’ duration of greater than 30 years. Patients who have no proteinuria after 20-25 years have a risk of developing overt renal disease of only approximately 1% per year.” (link). I’d note that a risk of 1% per year translates to a roughly 25% risk of developing overt renal disease over a 30 year time-frame, and that diabetics with the disease might not agree that a risk of that magnitude means that they are ‘unlikely’ to develop nephropathy, even if the annual risk is not high. Even if the annual risk were only half of that, 0,5%, the cumulative risk over a 30 year period would still be 14%, or roughly one in seven – are people with risks of that magnitude really ‘unlikely’ to develop nephropathy? This is certainly arguable. Many type 1 diabetics are diagnosed in childhood (peak incidence is in the early teenage years) and they can expect to live significantly longer than 20-25 years with the disease – if you disregard the ‘tail risk’ here, you seem in my opinion to be likely to neglect a substantial proportion of the total risk. This is incidentally not the only part of the book where I take issue with their coverage of topics related to diabetes, elsewhere in the book they note that:
“People who improve and maintain their fitness live longer […] Avoiding obesity helps too, but weight loss per se is only useful in reducing cardiovascular risk and the risk of developing diabetes when combined with regular exercise.”
Whereas in the case of nephropathy you can sort of argue about the language being imprecise and/or words meaning different things to different people, here things are a bit more clear because this is just plain WRONG. See e.g. Rana et al. (“Obesity and physical inactivity independently contribute to the development of type 2 diabetes; however, the magnitude of risk contributed by obesity is much greater than that imparted by lack of physical activity”). This is in my opinion the sort of error you should not find in a medical textbook.
Moving on to other parts of the coverage, let’s talk about angina. There are two types of angina – stable and unstable angina. Stable angina is induced by effort and relieved by rest. Unstable angina is angina of increasing severity or frequency, and it occurs at rest or minimal exertion. Unstable angina requires hospital admission and urgent treatment as it dramatically increases the risk of myocardial infarction. Some more stuff on related topics from the book:
“ACS [acute coronary syndrome] includes unstable angina and evolving MI [myocardial infarction], which share a common underlying pathology—plaque rupture, thrombosis, and inflammation”. Symptoms are: “Acute central chest pain, lasting >20min, often associated with nausea, sweatiness, dyspnoea [shortness of breath], palpitations [awareness of your heart beat]. May present without chest pain (‘silent’ infarct), eg in the elderly or diabetics. In such patients, presentations may include: syncope [fainting], pulmonary oedema, epigastric pain and vomiting, […] acute confusional state, stroke, and diabetic hyperglycaemic states.”
The two key questions to ask in the context of ACS are whether troponin (a cardiac enzyme) levels are elevated and whether there is ST-segment elevation. If there’s no ST-segment elevation and symptoms settle without a rise in troponin levels -> no myocardial damage (that’s the best case scenario – the alternatives are not as great..). In ACS, many deaths occur very soon after symptoms present; 50 % of deaths occur within two hours of symptom onset. “Up to 7% die before discharge.” Some MI complications have very high associated mortalities, for example a ventricular septal defect following an MI implies a 50% mortality rate during the first week alone.
Heart failure is a state in which the cardiac output is inadequate for the requirements of the body. It’s actually not that uncommon; the prevalence is 1-3% of the general population, increasing to roughly 10% “among elderly patients”. 25-50% die within 5 years of diagnosis, and if admission is needed the five year mortality rises to 75%.
Hypertension is a major risk factor for stroke and MI and according to the authors causes ~50% of all vascular deaths. Aside from malignant hypertension, which is relatively rare, hypertension is usually asymptomatic; the authors note specifically that “Headache is no more common than in the general population.” Isolated systolic hypertension, the most common form of hypertension, affects more than half of all people above the age of 60. “It is not benign: doubles risk of MI, triples risk of CVA [cerebrovascular accident, i.e. stroke].” The authors argue that: “Almost any adult over 50 would benefit from [antihypertensives], whatever their starting BP.” I think that’s downplaying the potential side effects of treatment, but it’s obvious that many people might benefit from treatment. Steps you can take to lower your BP without using medications according to the authors include: Reducing alcohol and salt intake, increasing exercise, reducing weight if obese, stop smoking, low-fat diet. They talk quite a bit about the different medications used to treat hypertension – I won’t cover that stuff in much detail, but I thought it was worth including the observation that ACE-inhibitors may be the 1st choice option in diabetics (especially if there’s renal involvement). On a related note, beta-blockers and thiazides may both increase the risk of new-onset diabetes.
No comments yet.