Reading a glucose curve — post-meal spike interpretation

Reading a post-meal glucose curve accurately is the single most actionable skill in personal diabetes management, and most people with diabetes have never been taught to do it. The curve generated by a CGM after a meal has a specific anatomy: a rise phase, a peak, and a return-to-baseline — and every feature of that shape carries information. A peak above 180 mg/dL at 45–60 minutes post-meal is a clinically significant spike. A peak that occurs at 90 minutes rather than 60 minutes suggests delayed gastric emptying (common with GLP-1 medications or gastroparesis). A glucose level that returns to baseline in under 90 minutes suggests the meal was well-matched to your insulin or residual pancreatic function. A glucose that peaks at 200 mg/dL, slowly drifts down, and is still 140 mg/dL at 3 hours suggests either a large carbohydrate load, insufficient insulin coverage, or both. Per the International Consensus on Time in Range (2019), the post-meal target for adults with Type 1 or Type 2 diabetes is glucose under 180 mg/dL — not the more aggressive 140 mg/dL target used in gestational diabetes, which reflects the tighter glucose requirements of pregnancy. This guide builds the analytical vocabulary needed to extract maximum information from post-meal CGM traces.

The anatomy of a normal post-meal glucose curve

In a metabolically healthy person without diabetes, a mixed meal triggers a predictable sequence. Glucose begins rising within 15–20 minutes of the first bite, driven by rapidly absorbed carbohydrates reaching the portal circulation. The curve peaks at approximately 110–130 mg/dL at 45–60 minutes post-meal, then declines as insulin-stimulated GLUT4 transporters move glucose into muscle and adipose cells. By 90–120 minutes, glucose is typically back at or near the pre-meal baseline. The area under this curve — the total glycemic excursion — is relatively small in people with intact beta-cell function because first-phase insulin secretion begins within 2–5 minutes of carbohydrate exposure, front-loading the response.¹

In Type 2 diabetes without medication, the reference anatomy shifts substantially. First-phase insulin secretion is blunted or absent, so glucose climbs higher and faster before second-phase secretion catches up. The typical peak in unmanaged Type 2 diabetes ranges from 180 to 250 mg/dL, occurring at 70–90 minutes rather than 45–60 minutes. The return to baseline is slower — often 3–4 hours rather than 2 — because insulin sensitivity in peripheral tissues is reduced. Even in well-controlled Type 2 diabetes on oral medication (metformin, SGLT2 inhibitors), post-meal peaks frequently exceed 180 mg/dL if the meal carbohydrate load is substantial, because these medications act on fasting glucose primarily, not on postprandial secretion.

In Type 1 diabetes, the curve shape depends almost entirely on the timing and dose of the mealtime bolus insulin. A correctly timed bolus (injected 15–20 minutes before eating for rapid-acting analogs like lispro or aspart) produces a curve that closely resembles the non-diabetic reference, with a modest peak and rapid return to range. A bolus given simultaneously with eating, or immediately after, produces a delayed insulin action that allows glucose to spike before the insulin concentration in tissue peaks — a textbook “delayed bolus” curve with a high peak and a later-than-expected drop.²

Understanding which reference curve applies to you is the starting point for all subsequent interpretation. A Type 1 patient reading their post-meal curve is primarily evaluating bolus timing and dose accuracy. A Type 2 patient is primarily evaluating the carbohydrate load and the adequacy of their medication regimen. The curve is the same shape but the diagnostic questions are different.

What peak height tells you — and what it doesn’t

A post-meal peak above 180 mg/dL is clinically significant by international consensus — it falls outside the target range and contributes to the time-above-range (TAR) metric that correlates with A1C and long-term complication risk.¹ But peak height is not the whole story. Two curves can share the same 200 mg/dL peak and have very different glycemic stress implications depending on their duration.

A sharp spike to 200 mg/dL that resolves to 100 mg/dL within 90 minutes represents a brief glucose excursion — high peak, short area under the curve (AUC). A flatter elevation to 155 mg/dL that persists for 4 hours represents a much larger total AUC despite a lower peak. Integrated glucose exposure (the AUC above a baseline threshold, typically 100 mg/dL) is the more complete measure of glycemic stress on tissue.³ Continuous glucose monitoring devices display glucose as a line over time; the area enclosed between that line and the baseline threshold is what drives advanced glycation end-product (AGE) formation in capillary endothelium, retinal cells, and renal glomeruli — the tissues most vulnerable to diabetic complications.

Practically, this means that a meal producing a brief 200 mg/dL spike from white rice that resolves in 80 minutes may be less harmful over a lifetime than a meal producing a sustained 160 mg/dL elevation from a high-fat, high-carbohydrate meal that keeps glucose elevated for 4 hours. Both patterns need attention; neither can be evaluated by peak height alone.

The clinical measure that captures this distinction is Time in Range (TIR) — the percentage of CGM readings between 70 and 180 mg/dL over a defined period, typically 14 days. The 2019 International Consensus recommends a TIR target of >70 % for adults with Type 1 or Type 2 diabetes. TIR below 70 % is associated with increased rates of microvascular complications independent of A1C, because A1C averages glucose over 3 months and cannot distinguish a high-peak, short-duration pattern from a low-peak, long-duration pattern that generates the same average.¹

Time-to-peak analysis — reading gastric emptying from your curve

The timing of the post-meal peak reveals information about how quickly food is leaving the stomach and entering the small intestine for absorption — a physiological process called gastric emptying that is frequently altered in diabetes and by common diabetes medications.

A peak arriving at 30–40 minutes post-meal typically indicates rapid gastric emptying combined with high glycemic index food: liquid glucose (juice, soda), white bread, white rice, or any meal eaten quickly. These foods reach the small intestine fast, produce a rapid rise in portal glucose, and generate the steep ascending limb characteristic of high-GI meals. For Type 1 patients, this rapid rise can outpace even a correctly timed bolus, requiring either a pre-meal bolus 20–30 minutes before eating or a shift to lower-GI alternatives.²

A peak consistently arriving at 90 minutes or later suggests delayed gastric emptying. In the context of diabetes, this can indicate three distinct causes: gastroparesis (autonomic neuropathy affecting the vagus nerve’s control of gastric motility, estimated to affect 30–50 % of people with long-standing Type 1 diabetes);⁴ GLP-1 receptor agonist use (semaglutide, liraglutide, dulaglutide all slow gastric emptying as part of their mechanism, which is largely beneficial but shifts peak timing); or simply high fat content in the meal, which physiologically delays emptying without any pathological basis. Distinguishing these causes requires considering the context — medication changes, meal composition — and if gastroparesis is suspected, formal gastric emptying scintigraphy.

For Type 1 patients with delayed gastric emptying, the standard bolus timing strategy becomes unreliable. Insulin injected 15 minutes before a meal may peak while glucose is still at pre-meal levels (because the meal hasn’t been absorbed yet), causing hypoglycemia in the first 90 minutes followed by a late glucose rise when absorption finally occurs 2–3 hours post-meal. Some endocrinologists recommend a post-meal bolus strategy (dosing after eating once the glucose trend is established) or using the CGM trend arrow to trigger a correction rather than a pre-meal bolus.

The “shoulder” pattern and dual-peak curves — what they mean

Not all post-meal glucose traces produce a clean single peak. Two characteristic abnormal patterns are worth recognizing: the shoulder and the dual peak.

A shoulder is a curve that rises to a primary peak, partially resolves, then plateaus at an elevated level rather than returning cleanly to baseline. This pattern often indicates that two overlapping food components are being absorbed at different rates — for example, a meal with both rapidly absorbed simple carbohydrates and slowly absorbed fat-modulated starches. The initial peak comes from the fast carbs; the shoulder reflects the slower second wave of absorption. High-fat restaurant meals (pizza, biryani with fatty meat) are common producers of this pattern. For Type 1 patients, dual-wave or extended bolus delivery (splitting the dose between an immediate bolus and a sustained delivery over 90 minutes) is the standard management strategy for these meals.

A dual-peak curve — a clear secondary rise 2–4 hours after the primary peak — has a different set of causes. In Type 1 diabetes, the most common explanation is the cephalic-phase insulin problem: the initial insulin bolus suppressed the primary glucose rise, but as bolus insulin concentration declines at 2–3 hours, a glucagon surge (or simply the tail of meal absorption) produces a secondary elevation. In patients with gastroparesis, a dual peak can represent food being delivered to the small intestine in two boluses as intermittent gastric contractions expel partially digested food. Identifying the pattern requires annotating the CGM trace with the meal time and comparing the timing across multiple similar meals to determine if it is reproducible.⁴

Using post-meal curves to identify problem foods

One of the most clinically valuable uses of a CGM is the structured food experiment: eating a specific food in isolation under controlled conditions and measuring the personal glucose response. This approach, used in the Weizmann Institute’s personalized nutrition research and the PREDICT study, reveals individual glycemic responses to specific foods that cannot be predicted from glycemic index tables alone.⁵

The protocol is straightforward. Choose a single food to test. Eat it in isolation after a 3-hour fast (to ensure pre-meal glucose is at baseline and no insulin from a prior meal is active). Eat a standardized portion (for example, 50 g of available carbohydrate — the amount used in official GI testing). Check CGM readings at 30, 60, 90, and 120 minutes. Record the peak value, the time to peak, and the return to baseline. Repeat the test on two separate occasions to confirm reproducibility.

Commonly tested comparisons that surprise people include: white rice versus brown rice (GI difference is real but smaller than expected — roughly 72 vs 55 — and the postprandial difference in a mixed meal is even smaller); fruit juice versus whole fruit (juice consistently produces faster, higher peaks because the fiber matrix that slows whole fruit absorption has been removed); sourdough bread versus standard white bread (sourdough’s lactic acid fermentation reduces starch digestibility, producing measurably lower glucose peaks despite comparable carbohydrate content);⁶ and cooled versus freshly cooked rice (cooling and reheating converts some starch to resistant starch, reducing glycemic impact by approximately 10–15 % in controlled studies).

Running 10 food experiments over 2 weeks produces a personalized glycemic map that is more accurate for that individual than any published GI table. The GI table represents a population average; the personal CGM trace represents you.

Sharing curve data with your endocrinologist — what to annotate

Raw CGM data without context is difficult for a clinician to interpret. An ambulatory glucose profile (AGP) report shows 14-day glucose patterns but does not tell the endocrinologist which meal caused which peak, whether the morning spike was food-driven or a dawn phenomenon, or whether the overnight hypoglycemia was post-exercise or inappropriate basal dosing.

The four data fields that most improve CGM review appointments are: (1) the time the meal started; (2) the approximate carbohydrate content in grams; (3) the insulin dose administered and its timing relative to the meal (for Type 1); and (4) any unusual circumstances — illness, exercise in the prior 24 hours, significant stress, unusual meal composition. These annotations turn a glucose trace from a curve into a data point that drives a specific clinical decision.

CalEye’s meal log timestamps are synchronized with clock time, which means that a CGM trace from a Dexcom or Libre device and a CalEye meal log from the same day can be visually overlaid — matching the meal log entry at 1:14 PM with the glucose rise beginning at 1:20 PM and peaking at 2:05 PM. This alignment makes it possible to present an endocrinologist with a paired record: meal composition on one side, glucose response on the other. The quality of the clinical conversation this enables is substantially higher than presenting either record alone.¹

Before each endocrinology appointment, export the past 14-day CGM report from your device’s companion app (Dexcom Clarity, LibreLink) and identify 3–5 representative post-meal traces — at least one that went well (peak under 180, returned to baseline in 2 hours) and at least one that did not. Annotate each with the four data fields above and bring the paired records to the appointment. The endocrinologist can then evaluate whether the poor post-meal trace represents an underdosed bolus, a miscounted carbohydrate load, an unusual food, or a timing error — and give you a specific rather than generic recommendation.

References

1. Battelino T, Danne T, Bergenstal RM, et al. “Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range.” Diabetes Care 42, no. 8 (2019): 1593–1603.

2. Heinemann L, Weyer C, Rauhaus M, Heinrichs S, Heise T. “Variability of the Metabolic Effect of Soluble Insulin and the Rapid-Acting Insulin Analog Insulin Aspart.” Diabetes Care 21, no. 11 (1998): 1910–1914.

3. Monnier L, Mas E, Ginet C, et al. “Activation of Oxidative Stress by Acute Glucose Fluctuations Compared With Sustained Chronic Hyperglycemia in Patients With Type 2 Diabetes.” JAMA 295, no. 14 (2006): 1681–1687.

4. Camilleri M, Bharucha AE, Farrugia G. “Epidemiology, Mechanisms, and Management of Diabetic Gastroparesis.” Clinical Gastroenterology and Hepatology 9, no. 1 (2011): 5–12.

5. Zeevi D, Korem T, Zmora N, et al. “Personalized Nutrition by Prediction of Glycemic Responses.” Cell 163, no. 5 (2015): 1079–1094.

6. Freitas DA, de Abreu LC, Valenti VE, et al. “Sourdough Bread and Glycemic Response: A Systematic Review.” International Journal of Food Sciences and Nutrition 71, no. 3 (2020): 295–304.