Romaine lettuce: eaten raw, occasionally risky, better with dressing

Most of what is firmly known about romaine is about germs. It is eaten raw, so nothing kills what it picks up in the field. In a 2011 E. coli outbreak across 10 US states, eating romaine raised the odds of falling ill tenfold. Of the 58 people who got sick, 67 percent ended up in hospital and 6.4 percent developed haemolytic uraemic syndrome, a serious complication that can damage the kidneys. Outbreaks are the exception. When the FDA tested 279 samples of raw romaine in 2020 and 2021, none carried Salmonella or the dangerous kinds of E. coli. On nutrition, the most useful finding is also the plainest. Seven adults ate a salad with romaine in it, and with a fat-free dressing they absorbed essentially none of its carotenoids, the yellow and orange pigments the body can turn into vitamin A. Oil in the dressing fixed that. Two things to know before reading further. We have no USDA entry for romaine in this cohort, so this page gives no nutrient table. And the blood pressure result below comes from lettuce fertilised on purpose to hold 530 mg of nitrate per 50 g. It was a test crop grown for the trial, and none of our cards measured how much nitrate ordinary romaine holds.

What your body absorbs

• In 7 adults eating a salad of spinach, romaine, tomatoes and carrots, a fat-free dressing gave almost no carotenoid absorption, while a 28 g oil dressing gave more than a 6 g one. — 7 adults, three salads with 0, 6 or 28 g canola oil dressing, 2-week washouts (n = 7) Evidence B source
  Negligible chylomicron carotenoids with 0 g fat; higher with 6 g (P<0.04) and higher still with 28 g (P<0.02)
• Vitamin K absorption from fresh romaine (179 µg) did not differ from spinach and broccoli, and raising meal fat from 30% to 45% of energy did not change it. — Adults aged 22-30 eating standard test meals (n = 3-8 per arm) Evidence C source
  No difference in phylloquinone absorption between romaine, spinach, broccoli, or 30 vs 45% fat
• NIH notes that vitamin K from vegetables is absorbed better when eaten with some fat, but still less than from oils. — General population Evidence E source
  Qualitative

What cooking and storage change

• Eating 250 g of fresh lettuce raised plasma antioxidant markers in 11 adults, but the same lettuce after 3 days in modified-atmosphere packaging raised none. — 11 healthy volunteers; lettuce type not stated as romaine (n = 11) Evidence C source
  TRAP AUC 1.3 fresh vs 0.1 stored mmol/L per 6 h; no marker changed after stored lettuce
• Pooled studies put Salmonella on about 4.1% of lettuce samples worldwide, and risk stays because lettuce is eaten raw. — Lettuce samples in published prevalence studies worldwide Evidence C source
  Salmonella prevalence 0.041 (95% CI 0.030-0.052); 0.028 developed vs 0.064 developing countries

What it does to your health

• Adding 100 g of lettuce to a high-fat meal lowered the 4-hour glucose curve (361.9 vs 385.7 mg/dL) and peak insulin in 16 healthy young men, with no change in blood fats. — 16 healthy young men, meals with 40% energy from fat; lettuce type not stated as romaine (n = 16) Evidence B source
  Glucose AUC 361.9 vs 385.7 control; peak insulin 33.5 vs 43.8 uIU/mL; TG, cholesterol, IL-6, TNF-alpha unchanged
• A test lettuce fertilised to hold 530 mg of nitrate per 50 g lowered systolic pressure by 4.9 mmHg within 3-6 hours compared with low-nitrate lettuce, in 19 young adults. — 19 healthy adults aged 22-31; lettuce type not stated as romaine (n = 19) Evidence B source
  SBP -4.9 mmHg (P = 0.031) between 3 and 6 h; plasma nitrate +240%

Safety, limits and interactions

• In a 2011 multistate E. coli O157 outbreak with 58 cases, eating romaine raised the odds of illness tenfold, 67% were hospitalised and 6.4% developed haemolytic uraemic syndrome. — 58 cases in 10 US states with matched controls (n = 58) Evidence C source
  Matched OR 10.0 (95% CI 2.1-97.0)
• The first US outbreak of E. coli O145, in 2010, was traced to bagged shredded romaine, with an odds ratio of 10.9 in Ohio. — 27 confirmed and 4 probable cases in five states (n = 31) Evidence C source
  Ohio OR 10.9 (95% CI 3.1-40.5); outbreak strain in an unopened bag
• Across 134 California-linked leafy green incidents from 1996 to 2016, romaine and spinach were the most often implicated foods. — US illness incidents linked to California leafy greens, about 2,240 confirmed cases (n = 134 incidents) Evidence C source
  Romaine and spinach most common vehicles; mostly pathogenic E. coli
• In FDA surveillance sampling of 279 raw romaine samples in 2020-2021, none carried Salmonella or disease-causing E. coli. — Romaine lettuce raw agricultural commodity samples, US (n = 279) Evidence E source
  0 of 279 positive for Salmonella or pathogenic E. coli

Who this works differently for

• NIH advises people on warfarin to keep vitamin K intake from foods such as leafy greens consistent, since sudden changes alter the drug's effect. — People taking warfarin and similar anticoagulants Evidence E source
  Qualitative

What is still unknown

• In a survey of 760 US adults, beliefs about the macronutrients in romaine and other vegetables mostly matched USDA values but differed by ethnicity, age and education. — 760 adults, online survey (n = 760) Evidence C source
  Perceptions broadly matched USDA; group differences significant

The bottom line

Put some fat on it. In the salad trial, 6 g of oil in the dressing already beat a fat-free one, and 28 g beat 6 g. That study had seven people, so treat the direction as likely and the exact amounts as rough. Vitamin K behaves a little differently. Romaine gave up its vitamin K as readily as spinach and broccoli did, and raising the fat in the meal from 30 to 45 percent of its energy changed nothing. NIH says vegetables release more vitamin K when eaten with some fat, though still less than oils do. So a normal meal with a little fat is enough, and drowning the salad buys you no extra vitamin K. Fresh seems to matter. Eleven adults ate 250 g of fresh lettuce and their blood antioxidant markers rose. The same lettuce after 3 days in a sealed modified-atmosphere bag moved none of them. The lettuce in that study, and in the two below, was not named as romaine. Adding 100 g of lettuce to a fatty meal lowered the 4-hour blood sugar curve in 16 healthy young men, from 385.7 to 361.9 mg/dL, and lowered peak insulin, with no change in blood fats. That was one meal in people without diabetes. Whether it holds over months, or in someone with diabetes, nobody in our cards has tested. The high nitrate test crop cut systolic blood pressure by 4.9 mmHg between 3 and 6 hours after eating, in 19 adults aged 22 to 31, compared with lettuce low in nitrate. That is a short effect in young healthy people and says nothing yet about long-term blood pressure. None of these findings comes from a meta-analysis, and none followed people for longer than a day. The safety side is broader than one outbreak. Across 134 illness incidents linked to California leafy greens between 1996 and 2016, romaine and spinach were the foods most often to blame. The first US outbreak of E. coli O145, in 2010, came from bagged shredded romaine, and the outbreak strain turned up in a bag nobody had opened, so a sealed bag is no proof of clean lettuce. Pooled studies find Salmonella on about 4.1 percent of lettuce samples worldwide, 2.8 percent in richer countries and 6.4 percent in poorer ones. If you take warfarin or a similar blood thinner, NIH advises keeping your vitamin K intake from leafy greens steady, because a sudden change up or down shifts how well the drug works. The advice is about keeping the amount steady, and it does not ask you to give greens up. Here is what our cards do not cover. None says who is most at risk from these infections, and none tests whether washing at home makes lettuce safer. We have nothing on typical portions, and nothing on popular claims about romaine. The one survey on beliefs found that 760 American adults mostly guessed its protein, fat and carbohydrate correctly, with differences by ethnicity, age and education.

Sources

1. rom-01 · randomised crossover trial
   The subjects (n = 7) each consumed 3 salads consisting of equivalent amounts of spinach, romaine lettuce, cherry tomatoes, and carrots with salad dressings containing 0, 6, or 28 g canola oil. ... Essentially no absorption of carotenoids was observed when salads with fat-free salad dressing were consumed. A substantially greater absorption of carotenoids was observed when salads were consumed with full-fat than with reduced-fat salad dressing.
   Am J Clin Nutr, 2004 · checked 2026-09-23
2. rom-04 · controlled clinical trial
   Absorption of phylloquinone from fresh spinach (165 microgram K), fresh broccoli (184 microgram K) and fresh romaine lettuce (179 microgram K) did not differ. There was no difference in phylloquinone absorption from fresh or cooked broccoli or from fresh romaine lettuce consumed with a meal containing 30 or 45% energy as fat.
   J Nutr, 1999 · checked 2026-09-23
3. rom-12 · fact sheet
   Consuming vegetables at the same time as some fat improves phylloquinone absorption from the vegetables, but the amount absorbed is still lower than that from oils.
   NIH Office of Dietary Supplements, Vitamin K Fact Sheet for Health Professionals · checked 2026-09-23
4. rom-05 · clinical trial
   Eleven healthy volunteers (six men, five women) consumed 250 g fresh lettuce ... No changes were observed after ingestion of MAP-stored lettuce for all the measured markers.
   Br J Nutr, 2002 · checked 2026-09-23
5. rom-09 · meta-analysis of prevalence
   the mean prevalence of Salmonella spp. on lettuce was 0.041 (95% CI: 0.030-0.052) ... Despite a relatively low prevalence, consumption of lettuce is inherently risky because it usually is eaten raw, without thermal treatment to inactivate pathogens.
   Food Microbiol, 2019 · checked 2026-09-23
6. rom-02 · randomised crossover trial
   Lettuce and watercress were added to the test meal in portions of 100 g ... The 0-4-h AUCs for glucose (385.7, 361.9, and 382.3 mg/dL for the control, lettuce, and watercress meals, respectively) were calculated to be lower when meal was consumed with lettuce compared to the control (P < 0.05) and watercress (P < 0.01) meals.
   Lipids Health Dis, 2021 · checked 2026-09-23
7. rom-03 · randomised crossover trial
   19 healthy participants (22-31 years) received 50 g of lettuce containing either 530 mg (8.4 mmol) nitrate + 11 mg (0.03 mmol) phenolic compounds (HNLP); or 3 mg nitrate (0.05 mmol) + 77 mg (0.2 mmol) phenolic compounds (LNHP), obtained by differential fertilisation. ... Systolic BP was reduced 4.9 ± 4.2 mmHg (P = 0.031) between 3 and 6 h after ingestion of HNLP compared with LNHP
   Eur J Nutr, 2022 · checked 2026-09-23
8. rom-06 · case-control
   We identified 58 cases in 10 states; 67% were hospitalized and 6.4% developed hemolytic uremic syndrome. Any romaine consumption was significantly associated with illness (matched Odds Ratio (mOR) = 10.0, 95% Confidence Interval (CI) = 2.1-97.0).
   PLoS One, 2013 · checked 2026-09-23
9. rom-07 · case-control
   illness was significantly associated with consumption of shredded romaine lettuce in Michigan (odds ratio [OR] = undefined; 95% confidence interval [CI] = 1.6 to undefined) and Ohio (OR = 10.9; 95% CI = 3.1 to 40.5). Samples from an unopened bag of shredded romaine lettuce yielded the predominant outbreak strain.
   J Food Prot, 2013 · checked 2026-09-23
10. rom-08 · systematic review
    In the 21-year period analyzed, 134 incidents were identified, the majority of which were surveillance related. ... Romaine lettuce and spinach were the most commonly implicated vehicles.
    J Food Prot, 2019 · checked 2026-09-23
11. rom-10 · agency surveillance
    The agency collected and tested a total of 279 samples for both pathogens. ... All the samples collected and tested for this assignment were negative for Salmonella spp. and strains of E. coli capable of causing serious illness.
    US FDA, Microbiological Surveillance Sampling FY20-21, Romaine Lettuce, 2022 · checked 2026-09-23
12. rom-11 · fact sheet
    People taking warfarin and similar anticoagulants need to maintain a consistent intake of vitamin K from food and supplements because sudden changes in vitamin K intakes can increase or decrease the anticoagulant effect [45].
    NIH Office of Dietary Supplements, Vitamin K Fact Sheet for Health Professionals · checked 2026-09-23
13. rom-13 · cross-sectional
    A total of 760 adults participated in the study. ... Participant perceptions of the macronutrient content of common vegetables and fruits largely coincided with the US Department of Agriculture values.
    Nutr Health, 2022 · checked 2026-09-23