CHAPTER 5 DIABETES COMPLICATIONS
In this section, specific complications associated with diabetes are discussed. Most of the mechanisms are vascular and the interventions will include components that reduce cardiovascular risk or improve glycaemic control. The assessment and interventions should be structured and follow agreed pathways.
Diabetes has long been associated with eye problems, including retinopathy, cataracts and glaucoma. Individually or collectively, these problems can lead to visual loss. Diabetic retinopathy is a specific vascular complication of diabetes and is the most common cause of newly registered cases of blindness (12%) in those aged 20 to 64 years in the UK (Evans et al 1991). In type 2 diabetics, glaucoma and cataracts are a more frequent cause of visual loss than retinopathy, but the latter is already present at diagnosis in one-fifth of patients (Macleod et al 1994) and its prevalence relates directly to the duration of diabetes.
Among the risk factors associated with the development and progression of diabetic retinal disease in type 2 diabetics are: raised blood pressure (UKPDS 1998c), increased duration of diabetes (Klein et al 1984), and the presence of microalbuminuria (Savage et al 1996).
The St Vincent Declaration of 1989 set a target to reduce new blindness caused by diabetes by one-third or more (Krans et al 1995). Regular accurate screening for diabetic eye disease can improve the prognosis by detecting abnormalities earlier in their natural history. Active management of microangiopathy affecting the retina has been possible since the development of laser photocoagulation. Cataract surgery and other surgical techniques that may benefit patients with diabetic eye disease are now available.
The two preferred methods of screening for diabetic retinopathy that meet the National Clinical Guidelines’ recommendation of a sensitivity of at least 80% and a specificity of at least 95% (Hutchinson et al 2001) are:
If a practice does not have access to either a slit-lamp or a retinal photography service, only suitably trained personnel should screen retinas using direct ophthalmoscopy (i.e. an ophthalmoscope) with mydriasis (pupil dilatation), although the reported sensitivity and specificity are lower for this method than the recommended minimum above and direct ophthalmoscopy risks missing clinically significant macular oedema (CSMO).
Retinopathy screening has been an early high priority in the Diabetes NSF: a stated aim was to offer it to 80% of diabetics by March 2006, rising to 100% by the end of 2007. At the time of writing, it is unclear how successful the NHS has been at meeting this aim. The data from the QOF reviews do not indicate which screening methods have been used.
Assessing visual acuity has limited value in detecting diabetic retinopathy, since not all retinopathy affects vision, especially if the macula is spared. When done, the clinician should use a properly illuminated Snellen chart of correct size for 3 or 6 metres and a pinhole to correct refractory errors.
Diabetic nephropathy is one of the more serious complications of diabetes. It occurs in 20–40% of patients with diabetes and its prevalence is increasing. Indo-Asian and Afro-Caribbean type 2 diabetics are at a greater risk of developing diabetic nephropathy than whites. In England, the rates for initiating treatment for end-stage renal disease were reported as 4.2 and 3.7 times higher in Afro-Caribbeans and Indo-Asians, respectively, than in whites (Roderick et al 1996).
Persistent albumin excretion greater than 300 mg/day, termed macroalbuminuria, in the absence of infection, marks the onset of diabetic nephropathy, and is also associated with increased mortality, particularly from vascular causes (Macleod et al 1995). The natural history of overt diabetic nephropathy is to progress to end-stage renal failure. The presence of lesser amounts of urinary albumin (less than 300 mg/day), termed microalbuminuria, indicates early renal damage. This is a risk factor for cardiovascular morbidity and mortality in type 2 diabetics (Dinneen & Gerstein 1997). About half of diabetics develop microalbuminuria at some stage. Of these (the study’s subjects were on insulin) 30–50% will progress to macroalbuminuria, but 30–50% will revert to normal albumin excretion and 20–30% will continue to have microalbuminuria (Laing et al 2003). If retinopathy is also present, then diabetes-related renal disease is the likely cause of the albuminuria. If retinopathy is absent, other causes of renal disease should be considered (NICE 2002b).
Prompt and effective interventions, particularly at an early stage, may both prevent end-stage renal failure and reduce the high risk of cardiovascular disease. The management of chronic kidney disease is now the subject of an NSF published in 2005 (DH Renal NSF Team 2006).
|Category||Spot collection (ng/ml creatinine)|
|Macro- (clinical) albuminuria||300 or more|
The “gold standard” for the measurement of urinary albumin excretion is a timed urine sample (either over 24 hours, enabling simultaneous measurement of creatinine clearance, or 4 hours), but this is not a practical screening procedure for widespread use in the community.
Sending an early morning urine sample (also referred to as a “spot check”) to a chemical pathology laboratory to measure the albumin:creatinine ratio (ACR) is both more practical and the preferred method recommended by the ADA. ACR levels equal to or greater than 2.5 mg/mmol in males or 3.5 mg/mmol in females indicate microalbuminuria. The test needs to be repeated for confirmation, as microalbuminuria can be transient. One of the authors completed a study in which the prevalence of microalbuminuria in diabetics in an ethnically mixed UK community was 19.3%. The characteristics independently associated with a higher prevalence of microalbuminuria were current insulin use, current smoking, older age, higher systolic blood pressure and poorer metabolic control, but there was no significant association with either increasing duration or gender. Unfortunately, the sample size was not large enough to determine whether there was an association between microalbuminuria and Indo-Asian ethnicity (Levene et al 2004).
From 2006 the reporting of estimated glomerular filtration rate (eGFR) has replaced the measurement of serum creatinine as standard in most laboratories. This change follows the recommendation of the Renal NSF. Serum creatinine levels may be affected by several factors, including age, gender, ethnicity, muscle mass, diet and some medications. Although not perfect, the eGFR is believed to be a better measure of renal function and easier for patients to understand than a serum creatinine result.
The glomerular filtration rate can be calculated using a formula and has been proposed as a more accurate, but logistically feasible, estimation of renal function. The 4-variable Modification of Diet and Renal Disease (MDRD) (Levey et al 1999) formula uses the patient’s creatinine, age (valid only between 18 and 70 years), sex and race:
Pathology laboratories have started reporting eGFR automatically (with white ethnicity as default) using the MDRD formula: the result must be multiplied by 1.21 if the patient is black African. If the laboratory does not provide an eGFR result or if looking at a historical creatinine result, then eGFR can be calculated using a readily available online calculator: www.kidney.org/professionals/kdoqi/gfr_calculator.cfm.
Since glomerular filtration rate is affected by the body’s surface area and the eGFR’s calculation does not include this, an eGFR result can underestimate renal function in large people and underestimate it in slim small people.
|1||≥ 90||Haematuria or proteinuria must be present|
|2||60–89||Haematuria or proteinuria must be present|
|5||≤ 15 or renal replacement therapy|
The UKPDS found that intensive diabetes management with the goal of achieving near normoglycemia delayed the onset of microalbuminuria and the progression of micro- to macroalbuminuria in type 2 diabetics (UKPDS 1998a, b).
Tight blood pressure control is the most important intervention for delaying the development of diabetic nephropathy (UKPDS 1998c). The target is a diastolic blood pressure below 75 mmHg in “normotensive insulin-dependent” or 90 mmHg in “hypertensive non-insulin-dependent” (BNF).
There is strong evidence for the reno-protective effect of ACE inhibitors/ARBs which have a greater effect upon urinary albumin excretion, a stronger marker for renal disease and cardiovascular health, than other blood-pressure-lowering drug classes. A Cochrane review suggests that ACE inhibitors are the best drug class to prevent microalbuminuria and nephropathy (Strippoli et al 2005). One randomised controlled trial (RCT), whose subjects were people with type 2 diabetes, hypertension, and microalbuminuria, reported that, compared with placebo, an ARB (irbesartan) reduced progression from early to late nephropathy over 2 years (Parving et al 2001). Another RCT, whose subjects were people with type 2 diabetes and early nephropathy found no significant difference in glomerular filtration rate change, mortality, or cardiovascular disease (CVD) events between an ARB (telmisartan) and an ACE inhibitor (enalapril) over 5 years (Barnett et al 2004). Other studies have found ARBs to be reno-protective in patients with type 2 diabetes and overt nephropathy (Brenner et al 2001, Lewis et al 2001).
However, effective blood pressure control is paramount, and non-DCCBs, β-blockers, or diuretics should be used in combination with or, if ACE inhibitor/ARB is contraindicated or not tolerated, as alternatives to inhibition of the renin-angiotensin-aldosterone system (Black et al 2003, Pepine et al 2003).
The small Danish Steno-2 study compared the effect of a targeted, intensified, multifactorial (both lifestyle and pharmacological) intervention with that of conventional treatment on modifiable risk factors for CVD in type 2 diabetics and microalbuminuria, showing significantly reduced hazard ratios for the development of both macro- and microvascular complications in type 2 diabetics with microalbuminuria (Gaede et al 2003).
Abnormal potassium levels can occur for a variety of reasons in patients with diabetes, and with or without nephropathy. Hyperkalaemia (serum potassium above the upper reference limit, usually 5.5 mEq/l) can result from ACE inhibitors, ARBs or potassium-sparing diuretics, and may occur in renal impairment. Hypokalaemia (serum potassium below the lower reference limit, usually 3.5 mEq/l) can result from long-term thiazide treatment without potassium-sparing medication or supplement. Treatment aims to restore normokalaemia.
Peripheral neuropathy causes lost sensation and autonomic dysfunction. Peripheral vascular disease (atherosclerosis of large and/or small leg vessels) causes ischaemia. Trauma (which may involve altered pressure-loading and be unnoticed by the patient) followed by infection complicates neuropathy and ischaemia to cause significant tissue damage, such as ulceration.
In both groups, the sequence of minor trauma, cutaneous ulceration and finally failure of the wound to heal can eventually lead to amputation, particularly if infection sets in, and rapid and significant tissue destruction occurs.
A UK population-based study in type 2 patients gave a prevalence of 1.4% for foot ulcers, but the prevalence of the risk factors that give rise to ulcers was 41.6% (Kumar et al 1994). Once a limb has been amputated the prognosis for the contralateral limb is poor (Ebskov & Josephsen 1980). Foot ulcers occur more frequently in whites than in Indo-Asians or Afro-Caribbeans, and are associated with adverse social circumstances such as deprivation and isolation (Boulton 1997), poor glycaemic control, the presence of other vascular risk factors (e.g. smoking) and increased duration of diabetes.