Insulin Isophane

What it is

Insulin isophane is ordinary human insulin, made in bacteria by recombinant DNA technology, crystallised together with protamine sulfate and zinc so that it dissolves slowly under the skin. That slow release is why it is called intermediate-acting: it has a slower onset and a longer duration than regular human insulin. It is a cloudy white suspension that must be mixed before use and injected under the skin only - never into a vein and never in a pump. Its amino acid sequence is identical to human insulin.

What the research says

Insulin replaces or supplements the body's own insulin, lowering blood glucose by pushing glucose into muscle and fat and switching off glucose production by the liver. The outcome evidence for insulin therapy is strong on microvascular complications: in type 1 diabetes intensive insulin reduced new retinopathy by 76%, and in newly diagnosed type 2 diabetes an intensive policy using sulphonylurea or insulin cut microvascular endpoints by 25%. Both bought that with more hypoglycaemia and more weight gain. The modern question is not whether NPH works but whether the newer long-acting analogues are meaningfully safer: Cochrane found fewer hypoglycaemic episodes with analogues but an absolute benefit of roughly one person in 100, and a 25,489-patient real-world study found no difference in hypoglycaemia emergencies at all.

Evidence grade: Well established.

How it works

Drug class: Intermediate-acting human insulin (protamine-crystallised insulin suspension)

Insulin lowers blood glucose in two ways at once: it makes skeletal muscle and fat take glucose out of the blood, and it stops the liver making new glucose. It also blocks the breakdown of fat and protein and increases protein building. The protamine in NPH slows absorption from the injection site so one injection covers several hours, with the peak effect at a median of about 6.5 hours but ranging from under 3 to 13 hours between people. (Source 1)

What it is used for

• Insulin is not optional in type 1 diabetes - Cochrane states people with type 1 diabetes need insulin for survival. The DCCT showed intensive insulin (3 or more injections daily or a pump) versus conventional 1 to 2 daily injections reduced new retinopathy by 76% and clinical neuropathy by 60%, at the cost of a two-to-threefold rise in severe hypoglycaemia. NPH remains a standard basal insulin, with modestly less severe hypoglycaemia reported for detemir and glargine. Evidence: established. (Source 2)
• In 3867 newly diagnosed patients, an intensive glucose-lowering policy using a sulphonylurea or insulin reduced any diabetes-related endpoint by 12% and microvascular endpoints by 25% over 10 years, with no significant effect on macrovascular disease. Major hypoglycaemia ran at 1.8% per year on insulin versus 0.7% on conventional treatment, and the insulin group gained 4.0 kg. Added to oral agents, NPH reached the same HbA1c as glargine (6.97% vs 6.96%) with more nocturnal hypoglycaemia. Evidence: established. (Source 3)
• The US label's indication is diabetes mellitus generally, so use in pregnancy is within the label. It states that decades of published data have not established an association between human insulin use in pregnancy and major birth defects, miscarriage, or adverse maternal or fetal outcomes, while poorly controlled diabetes itself carries a 6-10% background risk of major birth defects rising to 20-25% when HbA1c exceeds 10%. Evidence: established. (Source 4)

Interactions

• Alcohol (label): Alcohol can push blood glucose in either direction, and the label groups it with beta-blockers, clonidine and lithium as drugs that may increase or decrease insulin's glucose-lowering effect, needing dose adjustment and more glucose monitoring. (Source 5)
• Alcohol in the evening (delayed hypoglycaemia) (pharmacokinetic study): In a controlled crossover study in six men with type 1 diabetes, who were given regular insulin before meals plus a continuous basal insulin infusion rather than insulin isophane, wine in the evening did not change overnight glucose but was followed by hypoglycaemia the next morning in five of the six, with none after mineral water. The risk is delayed, not immediate. Six people is far too few to estimate how common this is, and the study says nothing about isophane insulin specifically. (Source 6)
• Niacin (nicotinic acid) supplements (clinical trial): Niacin raises blood glucose and so works against insulin. The label lists niacin among drugs that decrease insulin's glucose-lowering effect; a placebo-controlled trial found 1500 mg/d of extended-release niacin raised HbA1c from 7.2% to 7.5% in type 2 diabetes. (Source 7)
• Alpha-lipoic acid supplements (case reports): Alpha-lipoic acid is a recognised trigger of insulin autoimmune syndrome, in which insulin autoantibodies cause unpredictable hypoglycaemia. For someone already injecting insulin, an extra unexplained source of hypoglycaemia matters. (Source 8)
• Potassium-lowering medicines and potassium balance (label): Insulin shifts potassium from the blood into cells. On top of a potassium-lowering drug this can produce hypokalaemia, which untreated can cause respiratory paralysis, arrhythmia and death, so the label asks for potassium monitoring in people at risk. (Source 9)
• Beta-blockers, clonidine, guanethidine, reserpine (label): These blunt the warning signs of hypoglycaemia - the sweating, tremor and palpitations that normally prompt someone to eat - so a low glucose can go unnoticed until it is severe. (Source 5)
• Corticosteroids, thiazide and other diuretics, atypical antipsychotics, thyroid hormone, oral contraceptives, protease inhibitors (label): All of these raise blood glucose, so insulin requirements go up when they are started and down when they are stopped. The label asks for dose adjustment and more frequent glucose monitoring. (Source 5)
• ACE inhibitors, fibrates, fluoxetine, MAO inhibitors, salicylates, sulfonamide antibiotics, octreotide, pramlintide (label): These add to insulin's glucose-lowering effect and so increase the risk of hypoglycaemia; the label asks for dose adjustment and more frequent monitoring. (Source 5)
• Thiazolidinediones (pioglitazone, rosiglitazone) (label): Combined with insulin they cause dose-related fluid retention that can lead to or worsen heart failure, so the label asks for observation for signs of heart failure. (Source 9)
• Meals: timing and macronutrient content (label): NPH is injected before food is eaten, so a delayed, skipped or much smaller meal leaves insulin acting with nothing to act on. The label lists changes in meal pattern - macronutrient content or timing - among the reasons a dose needs adjusting. (Source 10)

Stopping it

• In type 1 diabetes insulin cannot be stopped. Cochrane states it plainly: it is needed for survival, which is why omitting it leads to diabetic ketoacidosis rather than simply higher glucose. (Source 2)
• Changing or stopping an insulin regimen is itself a risk point. The label says changes in insulin strength, manufacturer, type, injection site or method may affect control and predispose to hypoglycaemia or hyperglycaemia, and should be made under close supervision with more frequent monitoring. (Source 11)
• In older adults with type 2 diabetes there is a recognised case for going the other way - reducing or stopping glucose-lowering drugs. An international consensus review notes that most guidelines only address adding medication and few address reducing the burden, a concept known as deprescribing. (Source 12)
• That consensus describes what deprescribing means in practice: stopping high-risk medicines, lowering the dose, or substituting a less harmful agent - not an abrupt halt. (Source 12)
• There is no insulin withdrawal syndrome in the sense used for opioids or benzodiazepines. What happens on stopping is loss of glucose control, and the label's account of overdose shows the mirror image: too much insulin causes hypoglycaemia and hypokalaemia that can recur after apparent recovery. (Source 13)

What goes wrong

The price of intensive insulin therapy in the DCCT was a two-to-threefold increase in severe hypoglycaemia. (Source 14)

• Randomized trial, High certainty.
• Size: 1441 patients.
• Who: type 1 diabetes.
• How long: mean 6.5 years.
• Result: two-to-threefold increase in severe hypoglycaemia in the intensive arm.
• Funding: publicly funded.

The chief adverse event associated with intensive therapy was a two-to-threefold increase in severe hypoglycemia.

Insulin produced the highest rate of major hypoglycaemia of all the glucose-lowering policies in UKPDS and the greatest weight gain. (Source 3)

• Randomized trial, High certainty.
• Size: 3867 patients.
• Who: newly diagnosed type 2 diabetes.
• How long: 10 years.
• Result: major hypoglycaemic episodes per year: 0.7% conventional, 1.0% chlorpropamide, 1.4% glibenclamide, 1.8% insulin; weight gain 4.0 kg on insulin vs 2.6 kg chlorpropamide and 1.7 kg glibenclamide.
• Funding: multiple public and industry sources (not stated in the abstract)

The rates of major hypoglycaemic episodes per year were 0.7% with conventional treatment, 1.0% with chlorpropamide, 1.4% with glibenclamide, and 1.8% with insulin.

The intensive-control arms gained significantly more weight than conventional treatment, with insulin the worst. (Source 3)

• Randomized trial, High certainty.
• Size: 3867 patients.
• Who: newly diagnosed type 2 diabetes.
• How long: 10 years.
• Result: mean 2.9 kg gain in the intensive group versus conventional (p<0.001); 4.0 kg in those assigned insulin.
• Funding: multiple public and industry sources (not stated in the abstract)

Weight gain was significantly higher in the intensive group (mean 2.9 kg) than in the conventional group (p<0.001), and patients assigned insulin had a greater gain in weight (4.0 kg) than those assigned chlorpropamide (2.6 kg) or glibenclamide (1.7 kg).

Hypoglycaemia is the most common adverse reaction to insulin and severe episodes can cause seizures or death. (Source 11)

• Official position, Certainty not rated.
• Size: not stated.
• Who: people using insulin, including NPH.
• How long: not stated.
• Result: no frequency given in the label; awareness of hypoglycaemia is reduced in longstanding diabetes, neuropathy and with beta-blockers.
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

Hypoglycemia is the most common adverse reaction associated with insulins, including HUMULIN N. Severe hypoglycemia can cause seizures, may be life-threatening or cause death.

Insulin drives potassium into cells and can cause hypokalaemia, which untreated can be fatal. (Source 9)

• Official position, Certainty not rated.
• Size: not stated.
• Who: people using insulin, including NPH.
• How long: not stated.
• Result: no frequency given; the label advises monitoring potassium in people on potassium-lowering medicines.
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

All insulins, including HUMULIN N, cause a shift in potassium from the extracellular to intracellular space, possibly leading to hypokalemia. Untreated hypokalemia may cause respiratory paralysis, ventricular arrhythmia, and death. Monitor potassium levels in patients at risk for hypokalemia if indicated (e.g., patients using potassium-lowering medications, patients taking medications sensitive to serum potassium concentrations).

Repeated injection into the same site causes fat loss or fat thickening, and localised skin amyloidosis, which in turn makes absorption erratic. (Source 15)

• Official position, Certainty not rated.
• Size: postmarketing reports of uncertain denominator.
• Who: people injecting insulin subcutaneously.
• How long: not stated.
• Result: frequency cannot be reliably estimated; hyperglycaemia reported when injecting into amyloidosis, hypoglycaemia on moving to an unaffected site.
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

Localized cutaneous amyloidosis at the injection site has occurred. Hyperglycemia has been reported with repeated insulin injections into areas of localized cutaneous amyloidosis; hypoglycemia has been reported with a sudden change to an unaffected injection site.

Weight gain is an expected consequence of insulin treatment, attributed to its anabolic effects and less glucose lost in urine. (Source 15)

• Official position, Certainty not rated.
• Size: not stated.
• Who: people using insulin, including NPH.
• How long: not stated.
• Result: no frequency given in the label (UKPDS measured 4.0 kg over 10 years in the insulin arm)
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

Weight gain has occurred with insulins, including HUMULIN N, and has been attributed to the anabolic effects of insulin and the decrease in glycosuria.

The timing of NPH's peak effect varies several-fold between people, which is why its action cannot be predicted precisely. (Source 1)

• Blood level study, Moderate certainty.
• Size: 16 healthy subjects, 4 occasions each.
• Who: healthy volunteers given 0.4 unit/kg subcutaneously.
• How long: single doses.
• Result: median maximum effect at 6.5 hours, range 2.8 to 13 hours; median peak serum insulin at about 4 hours, range 1 to 12 hours.
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

In a study in which healthy subjects (n=16) received subcutaneous injections of HUMULIN N (0.4 unit/kg) on 4 occasions, the median maximum effect occurred at 6.5 hours (range: 2.8 to 13 hours).

Insulin combined with a thiazolidinedione can cause dose-related fluid retention that may lead to or worsen heart failure. (Source 9)

• Official position, Certainty not rated.
• Size: not stated.
• Who: people taking insulin plus a PPAR-gamma agonist.
• How long: not stated.
• Result: no frequency given in the label.
• Funding: manufacturer label (Eli Lilly), revised 12/2025.

Thiazolidinediones (TZDs), which are peroxisome proliferator-activated receptor (PPAR)-gamma agonists, can cause dose-related fluid retention, when used in combination with insulin. Fluid retention may lead to or exacerbate heart failure.

Evening alcohol was followed by delayed next-morning hypoglycaemia in type 1 diabetes treated with regular insulin plus a basal insulin infusion, in a controlled crossover study. (Source 6)

• Blood level study, Low certainty.
• Size: 6 men with type 1 diabetes.
• Who: aged 19 to 51, HbA1c 7.0 to 10.3%, on regular insulin plus basal infusion.
• How long: two overnight admissions, 5 pm to noon next day.
• Result: postprandial peak glucose 8.9 vs 15 mmol/L (P<0.01); 5 of 6 subjects needed treatment for hypoglycaemia next morning (nadir 1.9-2.9 mmol/L) after wine and none after water.
• Funding: not stated.

Limit of this finding: This was a crossover experiment in six men with type 1 diabetes, all studied on a ward. It shows the effect can happen and describes a mechanism; it is far too small to say how often it happens, and it included no women, so the five-in-six figure should not be read as a population rate. The insulin used was regular (soluble) insulin before meals plus a continuous basal insulin infusion at 0.15 mU/kg/min, not insulin isophane, so this describes insulin treatment in general and is not evidence about isophane insulin specifically.

In the morning, fasting and postprandial blood glucose levels were significantly lower after consumption of wine (postprandial peak 8.9 [1.7] vs. 15 [1.5] mmol/l, P < 0.01), and from 10:00 A.M., five subjects required treatment for hypoglycemia (nadir 1.9-2.9 mmol/l). None of the subjects had hypoglycemia after consumption of water.

Extended-release niacin at 1500 mg/d worsened HbA1c in people with type 2 diabetes in a placebo-controlled trial. (Source 7)

• Randomized trial, Moderate certainty.
• Size: 148 patients randomised.
• Who: adults with type 2 diabetes and diabetic dyslipidaemia, 47% also on a statin.
• How long: 16 weeks.
• Result: HbA1c 7.2% to 7.5% on 1500 mg/d niacin versus 7.13% to 7.11% on placebo (P=0.048); 1000 mg/d not significant (P=0.16)
• Funding: trial of a branded extended-release niacin product (Niaspan); funding not stated in the abstract.

Baseline and week 16 values for glycosylated hemoglobin levels were 7.13% and 7.11%, respectively, in the placebo group; 7.28% and 7.35%, respectively, in the 1000-mg ER niacin group (P=.16 vs placebo); and 7.2% and 7.5%, respectively, in the 1500-mg ER niacin group (P=.048 vs placebo).

Alpha-lipoic acid supplements can trigger insulin autoimmune syndrome with severe hypoglycaemia. (Source 8)

• Case report, Very low certainty.
• Size: 2 cases plus a literature review.
• Who: Indian patients taking alpha-lipoic acid nutraceuticals.
• How long: weeks from starting the supplement.
• Result: one case needed oral steroids tapered over months; the other remitted with dietary measures alone.
• Funding: not stated.

In the first case, rapid-onset severe hypoglycemia mandated the use of oral steroids with gradual tapering and stoppage over a duration of few months, while in the second case, mild hypoglycemia was essentially managed with dietary modifications alone

The label restricts NPH to subcutaneous injection and forbids intravenous use and use in an insulin infusion pump. (Source 16)

• Official position, Certainty not rated.
• Size: not applicable - regulatory position.
• Who: anyone prescribed HUMULIN N.
• How long: not applicable.
• Result: subcutaneous route only, into abdominal wall, thigh, upper arm or buttocks, with sites rotated; intravenous administration and insulin infusion pump use are prohibited.
• Funding: not applicable - FDA-approved labelling.

Do not administer HUMULIN N intravenously and do not use HUMULIN N in an insulin infusion pump.

What the evidence supports

Intensive insulin therapy in type 1 diabetes cut the development of retinopathy by 76% compared with conventional one-or-two-injection treatment. (Source 14)

• Randomized trial, High certainty.
• Size: 1441 patients (726 primary prevention, 715 secondary intervention)
• Who: insulin-dependent (type 1) diabetes, ages 13 to 39.
• How long: mean 6.5 years.
• Result: retinopathy development risk reduced 76% (95% CI 62 to 85%); progression slowed 54% (39 to 66%); microalbuminuria reduced 39% (21 to 52%); clinical neuropathy reduced 60% (38 to 74%)
• Funding: publicly funded (National Institute of Diabetes and Digestive and Kidney Diseases)

In the primary-prevention cohort, intensive therapy reduced the adjusted mean risk for the development of retinopathy by 76 percent (95 percent confidence interval, 62 to 85 percent), as compared with conventional therapy.

In newly diagnosed type 2 diabetes, intensive glucose control with sulphonylurea or insulin reduced any diabetes-related endpoint by 12% and microvascular endpoints by 25%, but not macrovascular disease or death. (Source 3)

• Randomized trial, High certainty.
• Size: 3867 newly diagnosed patients.
• Who: median age 54, fasting plasma glucose 6.1-15.0 mmol/L after 3 months of diet.
• How long: 10 years.
• Result: any diabetes-related endpoint 12% lower (95% CI 1 to 21, p=0.029); microvascular 25% lower (7 to 40, p=0.0099); diabetes-related death 10% lower (-11 to 27, p=0.34); all-cause mortality 6% lower (-10 to 20, p=0.44); HbA1c 7.0% vs 7.9%.
• Funding: multiple public and industry sources (not stated in the abstract)

Over 10 years, haemoglobin A1c (HbA1c) was 7.0% (6.2-8.2) in the intensive group compared with 7.9% (6.9-8.8) in the conventional group--an 11% reduction. There was no difference in HbA1c among agents in the intensive group. Compared with the conventional group, the risk in the intensive group was 12% lower (95% CI 1-21, p=0.029) for any diabetes-related endpoint; 10% lower (-11 to 27, p=0.34) for any diabetes-related death; and 6% lower (-10 to 20, p=0.44) for all-cause mortality.

HbA1c fell slightly further with NPH than with analogues in that real-world cohort. (Source 17)

• Cohort study, Moderate certainty.
• Size: 25,489 patients.
• Who: adults with type 2 diabetes starting basal insulin.
• How long: 1 year after initiation.
• Result: 9.4% to 8.2% with analogues versus 9.4% to 7.9% with NPH; adjusted difference-in-differences -0.22% (95% CI -0.09 to -0.37)
• Funding: not stated in the abstract.

Within 1 year of insulin initiation, hemoglobin A1c level decreased from 9.4% (95% CI, 9.3% to 9.5%) to 8.2% (95% CI, 8.1% to 8.2%) after initiation of insulin analogs and from 9.4% (95% CI, 9.3% to 9.5%) to 7.9% (95% CI, 7.9% to 8.0%) after initiation of NPH insulin

What the evidence does not support

HbA1c achieved with NPH was comparable to that achieved with long-acting analogues in type 2 diabetes. (Source 18)

• Systematic review, Low certainty.
• Size: 24 RCTs, 4740 participants on analogues plus NPH comparators.
• Who: adults with type 2 diabetes.
• How long: 24 weeks to 5 years.
• Result: no meaningful difference in HbA1c change; no clear difference in weight gain.
• Funding: trials largely industry-sponsored.

Overall, insulin glargine and insulin detemir resulted in fewer participants experiencing hypoglycaemia when compared with NPH insulin. Changes in HbA1c were comparable for long-acting insulin analogues and NPH insulin.

Glargine did not significantly reduce severe hypoglycaemia compared with NPH in type 1 diabetes. (Source 2)

• Systematic review, Moderate certainty.
• Size: 9 studies, 2350 participants.
• Who: adults and children with type 1 diabetes.
• How long: 24 to 104 weeks.
• Result: 122/1191 (10.2%) glargine vs 145/1159 (12.5%) NPH, RR 0.84 (95% CI 0.67 to 1.04); HbA1c mean difference 0.02%.
• Funding: clinical study reports obtained from pharmaceutical companies and regulators.

Severe hypoglycaemia was observed in 122/1191 participants (10.2%) in the insulin glargine group compared with 145/1159 participants (12.5%) in the NPH insulin group (RR 0.84, 95% CI 0.67 to 1.04; 9 studies, 2350 participants; moderate-certainty evidence).

In routine care, starting a basal analogue instead of NPH was not associated with fewer hypoglycaemia emergency visits or better glycaemic control. (Source 17)

• Cohort study, Moderate certainty.
• Size: 25,489 patients (1928 analogue, 23,561 NPH); 4428 propensity-matched.
• Who: adults with type 2 diabetes starting basal insulin, Kaiser Permanente Northern California 2006-2015.
• How long: mean follow-up 1.7 years.
• Result: 11.9 (95% CI 8.1 to 15.6) vs 8.8 (7.9 to 9.8) hypoglycaemia-related ED visits or admissions per 1000 person-years; between-group difference 3.1 (95% CI -1.5 to 7.7), P=0.07; matched adjusted HR 1.16 (0.71 to 1.78)
• Funding: not stated in the abstract.

During a mean follow-up of 1.7 years, there were 39 hypoglycemia-related ED visits or hospital admissions among 1928 patients who initiated insulin analogs (11.9 events [95% CI, 8.1 to 15.6] per 1000 person-years) compared with 354 hypoglycemia-related ED visits or hospital admissions among 23 561 patients who initiated NPH insulin (8.8 events [95% CI, 7.9 to 9.8] per 1000 person-years) (between-group difference, 3.1 events [95% CI, -1.5 to 7.7] per 1000 person-years; P = .07).

Where the evidence is mixed

Cochrane found glargine and detemir caused fewer hypoglycaemic episodes than NPH in type 2 diabetes, but the absolute benefit was about one person in 100 and the certainty very low to low. (Source 18)

• Systematic review, Very low certainty.
• Size: 24 RCTs; 3419 randomised to glargine, 1321 to detemir, compared with NPH.
• Who: adults with type 2 diabetes.
• How long: 24 weeks to 5 years.
• Result: glargine vs NPH severe hypoglycaemia RR 0.68 (95% CI 0.46 to 1.01), absolute risk reduction -1.2% (95% CI -2.0 to 0); detemir vs NPH serious hypoglycaemia Peto OR 0.16 (0.04 to 0.61), ARR -0.9% (-1.1 to -0.4)
• Funding: trials largely industry-sponsored; all trials had unclear or high risk of bias in several domains.

However, serious hypoglycaemic events were rare and the absolute risk reducing effect was low. Approximately one in 100 people treated with insulin detemir instead of NPH insulin benefited. In the studies, low blood glucose and HbA1c targets, corresponding to near normal or even non-diabetic blood glucose levels, were set. Therefore, results from the studies are only applicable to people in whom such low blood glucose concentrations are targeted.

Cochrane reports that the trials comparing analogues with NPH almost never measured the outcomes patients care about most. (Source 18)

• Systematic review, Low certainty.
• Size: 24 RCTs.
• Who: adults with type 2 diabetes.
• How long: 24 weeks to 5 years.
• Result: death, diabetes complications, quality of life and socioeconomic effects insufficient or lacking in almost all trials.
• Funding: trials largely industry-sponsored.

Information on patient-relevant outcomes such as death from any cause, diabetes-related complications, health-related quality of life and socioeconomic effects was insufficient or lacking in almost all included trials.

In type 1 diabetes, severe hypoglycaemia occurred in 11.5% on NPH versus 8.5% on detemir, a moderate-certainty difference whose prediction interval crossed no effect. (Source 2)

• Systematic review, Moderate certainty.
• Size: 8 studies, 3219 participants (within a review of 26 RCTs and 8784 randomised participants)
• Who: adults and children with type 1 diabetes.
• How long: 24 to 104 weeks.
• Result: 171/2019 (8.5%) detemir vs 138/1200 (11.5%) NPH, RR 0.69 (95% CI 0.52 to 0.92); 95% prediction interval 0.34 to 1.39.
• Funding: clinical study reports obtained from pharmaceutical companies and regulators.

There was a reduction in severe hypoglycaemia in favour of insulin detemir: 171/2019 participants (8.5%) in the insulin detemir group compared with 138/1200 participants (11.5%) in the NPH insulin group experienced severe hypoglycaemia (RR 0.69, 95% CI 0.52 to 0.92; 8 studies, 3219 participants; moderate-certainty evidence).

Added to oral agents and titrated to target, NPH reached the same HbA1c as glargine but with more nocturnal hypoglycaemia. (Source 19)

• Randomized trial, Moderate certainty.
• Size: 756 overweight adults.
• Who: type 2 diabetes with HbA1c above 7.5% on one or two oral agents.
• How long: 24 weeks.
• Result: end-point HbA1c 6.97% NPH vs 6.96% glargine; 26.7% vs 33.2% reached HbA1c 7% or less without documented nocturnal hypoglycaemia (P<0.05); other symptomatic hypoglycaemia 21-48% lower with glargine.
• Funding: trial of insulin glargine; funding not stated in the abstract.

However, nearly 25% more patients attained this without documented nocturnal hypoglycemia (<or=72 mg/dl [4.0 mmol/l]) with glargine (33.2 vs. 26.7%, P < 0.05). Moreover, rates of other categories of symptomatic hypoglycemia were 21-48% lower with glargine.

An overview of 11 systematic reviews found analogues reduced nocturnal hypoglycaemia versus NPH but made no difference to severe hypoglycaemia. (Source 20)

• Review of reviews, Moderate certainty.
• Size: 11 systematic reviews containing 25 clinical trials.
• Who: people with type 1 diabetes.
• How long: varied.
• Result: nocturnal hypoglycaemia RR 0.66 (95% CI 0.57 to 0.76); severe hypoglycaemia RR 0.94 (95% CI 0.71 to 1.24)
• Funding: not stated.

We found evidence that LAIA are efficacious compared to NPH, with estimates showing a reduction in nocturnal hypoglycemia episodes (RR 0.66; 95% CI 0.57; 0.76) and A1C (95% CI 0.23; 0.12). No significance was found related to severe hypoglycemia (RR 0.94; 95% CI 0.71; 1.24).

Where the research disagrees

Whether long-acting insulin analogues are meaningfully safer than NPH

• Cochrane review of 24 RCTs in type 2 diabetes (2020), meta-analysis of randomised trials, very low to low certainty, trials titrated to near-normal glucose targets: However, serious hypoglycaemic events were rare and the absolute risk reducing effect was low. Approximately one in 100 people treated with insulin detemir instead of NPH insulin benefited. (Source 18)
• Lipska and colleagues, JAMA (2018), 25,489 patients in routine care, retrospective observational cohort with propensity matching: Among patients with type 2 diabetes, initiation of a basal insulin analog compared with NPH insulin was not associated with a reduced risk of hypoglycemia-related ED visits or hospital admissions or with improved glycemic control. (Source 17)
• Overview of 11 systematic reviews in type 1 diabetes (2018), overview of systematic reviews with new meta-analysis: We found evidence that LAIA are efficacious compared to NPH, with estimates showing a reduction in nocturnal hypoglycemia episodes (RR 0.66; 95% CI 0.57; 0.76) and A1C (95% CI 0.23; 0.12). No significance was found related to severe hypoglycemia (RR 0.94; 95% CI 0.71; 1.24). (Source 20)

How much

• Reference intake: There is no fixed dose. As a position, the Humulin N label (Eli Lilly, revised 12/2025) states the dosage is individualised and adjusted by the prescriber according to the person's metabolic needs, blood glucose monitoring results and glycaemic goal, and that it must be injected subcutaneously only, never intravenously and never in a pump. (Source 10)
• Upper limit: The label sets no maximum dose, because insulin requirement varies by person. What it does state is the consequence of too much: excess insulin may cause hypoglycaemia and hypokalaemia, and hypoglycaemia can recur after apparent recovery so sustained carbohydrate intake and observation may be needed. Position of the Humulin N label, revised 12/2025. (Source 13)
• Studied: The label's pharmacodynamic and pharmacokinetic studies gave healthy subjects single subcutaneous doses of 0.4 unit/kg. (Source 1)
• Studied: In the DCCT, intensive therapy was an external insulin pump or three or more daily injections; conventional therapy was one or two daily injections. (Source 14)
• Studied: In the treat-to-target trial, bedtime NPH was titrated once daily against a fasting plasma glucose target of 100 mg/dL or less on top of existing oral agents. (Source 19)

A common belief, and what the research shows

The belief: NPH is an obsolete insulin, and the newer long-acting analogues are clearly safer.

What the research shows: The difference is real but small, and it shrinks outside trial conditions. Cochrane's own summary of 24 randomised trials in type 2 diabetes says: "However, serious hypoglycaemic events were rare and the absolute risk reducing effect was low. Approximately one in 100 people treated with insulin detemir instead of NPH insulin benefited." HbA1c was comparable and there was no clear difference in weight gain. In 25,489 people starting basal insulin in routine care, hypoglycaemia-related emergency visits were 11.9 per 1000 person-years on analogues and 8.8 on NPH, with no significant difference, and HbA1c fell slightly further on NPH. The trials that showed the hypoglycaemia advantage aimed at near-normal glucose targets that current guidelines do not recommend for most people.

Questions and answers

What is it?

Insulin isophane, usually called NPH, is human insulin made in bacteria and then crystallised with protamine sulfate so it dissolves slowly after injection. That makes it intermediate-acting: slower to start and longer-lasting than regular insulin. It is a cloudy suspension, injected under the skin only. (Source 21)

What does it do in the body?

It lowers blood glucose by making muscle and fat take glucose out of the blood and by stopping the liver from producing glucose. It also blocks the breakdown of fat and protein and increases protein building, which is part of why insulin treatment tends to cause weight gain. (Source 1)

Is it good or bad for you?

Good in the sense that few treatments have this much outcome evidence: intensive insulin in type 1 diabetes reduced new retinopathy by 76% and clinical neuropathy by 60%. Harmful in the sense that the same trial produced a two-to-threefold increase in severe hypoglycaemia, and UKPDS found more weight gain with insulin than with any other glucose-lowering policy. The balance depends on how tight the glucose target is and on who is being treated. (Source 14)

How do you get more of it?

This is not a question of getting more. NPH is prescription-only and the amount is individualised by the prescriber from metabolic needs, glucose readings and the glycaemic target, and adjusted when activity, meals, kidney or liver function, or illness change. It is injected subcutaneously, never into a vein and never in a pump. (Source 10)

If it is harmful, what reduces it?

Too much insulin cannot be removed once injected; it has to be countered with glucose. The label states that excess insulin may cause hypoglycaemia and hypokalaemia, that mild episodes can be treated with oral glucose and severe ones with glucagon or intravenous glucose, and that sustained carbohydrate intake and observation may be needed because hypoglycaemia can come back after apparent recovery. (Source 13)

Why might someone be low in it or missing it?

Someone needs injected insulin when their own pancreas cannot supply enough. In type 1 diabetes the insulin-producing cells are destroyed and insulin is needed for survival. In type 2 diabetes insulin is usually added later, when diet, metformin and other agents no longer hold glucose at target - the UKPDS and treat-to-target trials both studied exactly that step. (Source 2)

Which whole foods contain it or feed it?

No food contains injectable insulin. Food matters in the opposite direction: because NPH is already working under the skin, what and when someone eats determines whether the dose fits. The label lists changes in meal patterns - macronutrient content or timing of food intake - alongside changes in physical activity and illness as reasons the dose needs adjusting. (Source 10)

What happens if you do not have it?

In type 1 diabetes, going without insulin is not survivable - Cochrane states that people with type 1 diabetes need treatment with insulin for survival. In type 2 diabetes, insufficient glucose lowering over years shows up as microvascular damage: in UKPDS, tighter control cut microvascular endpoints including the need for retinal photocoagulation by 25% over 10 years. (Source 2)

How can you test for it?

There is no routine test of insulin level in someone being treated with insulin. What is measured is the effect: self-monitored blood glucose or continuous glucose monitoring day to day, and HbA1c over months. The label calls self-monitoring essential to preventing and managing hypoglycaemia and asks for more frequent checks in anyone at higher risk or with reduced awareness of lows. (Source 11)

References

1. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
2. Cochrane Database of Systematic Reviews. (Ultra-)long-acting insulin analogues for people with type 1 diabetes mellitus. 2021. PMID 33662147, DOI 10.1002/14651858.CD013498.pub2. Read the source
3. The Lancet. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). 1998. PMID 9742976, DOI 10.1016/S0140-6736(98)07019-6. Read the source
4. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
5. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
6. Diabetes Care. The effect of evening alcohol consumption on next-morning glucose control in type 1 diabetes. 2001. PMID 11679452, DOI 10.2337/diacare.24.11.1888. Read the source
7. Archives of Internal Medicine. Efficacy, safety, and tolerability of once-daily niacin for the treatment of dyslipidemia associated with type 2 diabetes: results of the assessment of diabetes control and evaluation of the efficacy of niaspan trial. 2002. PMID 12123399, DOI 10.1001/archinte.162.14.1568. Read the source
8. Journal of Postgraduate Medicine. Autoimmune hypoglycemia due to alpha-lipoic acid: Report of two cases. 2025. PMID 40488315, DOI 10.4103/jpgm.jpgm_58_25. Read the source
9. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
10. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
11. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
12. The Lancet Diabetes & Endocrinology. Realigning diabetes regimens in older adults: a 4S Pathway to guide simplification and deprescribing strategies. 2025. PMID 39978368, DOI 10.1016/S2213-8587(24)00372-3. Read the source
13. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
14. New England Journal of Medicine. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. 1993. PMID 8366922, DOI 10.1056/NEJM199309303291401. Read the source
15. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
16. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source
17. JAMA. Association of Initiation of Basal Insulin Analogs vs Neutral Protamine Hagedorn Insulin With Hypoglycemia-Related Emergency Department Visits or Hospital Admissions and With Glycemic Control in Patients With Type 2 Diabetes. 2018. PMID 29936529, DOI 10.1001/jama.2018.7993. Read the source
18. Cochrane Database of Systematic Reviews. (Ultra-)long-acting insulin analogues versus NPH insulin (human isophane insulin) for adults with type 2 diabetes mellitus. 2020. PMID 33166419, DOI 10.1002/14651858.CD005613.pub4. Read the source
19. Diabetes Care. The treat-to-target trial: randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. 2003. PMID 14578243, DOI 10.2337/diacare.26.11.3080. Read the source
20. PLoS ONE. Long-acting insulin analogues for type 1 diabetes: An overview of systematic reviews and meta-analysis of randomized controlled trials. 2018. PMID 29649221, DOI 10.1371/journal.pone.0194801. Read the source
21. DailyMed, U.S. National Library of Medicine. HUMULIN N (insulin isophane human) injectable suspension - FDA prescribing information (Eli Lilly and Company), label revised 12/2025. 2025. Read the source