A 68 year old retired physician consults you about an episode of severe dyspepsia on a background of 20 years of mild indigestion. She is overweight but otherwise in good health. She is taking omeprazole 20mg nocte, drinks 10 units of alcohol per week and takes naproxen for arthralgias two to three times per week. You perform an oesophago-gastro-duodenoscopy which reveals a small 5 cm hiatus hernia and mild antral gastritis but is otherwise normal. You take an antral biopsy and apply a CLO test for helicobacter pylori. This is strongly positive. Histological investigation confirms both gastritis and helicobacter infection.
You wonder if there is any new evidence about helicobacter eradication in patients with non-ulcer dyspepsia and formulate the question, In a patient with nonulcer dyspepsia and helicobacter pylori infection, will helicobacter eradication therapy result in a reduction in symptoms of dyspepsia?
You perform a MEDLINE search using the terms “nonulcer dyspepsia” and “helicobacter pylori” and “randomized controlled trial
And retrieve two RCTs that appear to come to conflicting conclusions
Read the article and decide:
- Is the evidence from this randomised trial valid?
- If valid, is this evidence important?
- If valid and important, can you apply this evidence in caring for your patient?
Then explain why the two trials come to different conclusions. What will you tell your patient?
Completed Therapy Worksheet I for Evidence-Based Gastroenterology and Hepatology
Citation
Symptomatic benefit from eradicating helicobacter pylori infection in patients with nonulcer dyspepsia. McColl K et al N Engl J Med 1998;339:1869-74.
Are the results of this single preventive or therapeutic trial valid?
- Was the assignment of patients to treatments randomised?
- And was the randomisation list concealed?
- Yes
- Were all patients who entered the trial accounted for at its conclusion?
- And were they analysed in the groups to which they were randomised?
- Yes
- Were patients and clinicians kept “blind” to which treatment was being received?
- Yes. The study was placebo controlled.
- Aside from the experimental treatment, were the groups treated equally?
- Yes
- Were the groups similar at the start of the trial?
- Yes
Are the valid results of this randomised trial important?
Sample Calculations
| Occurrence of symptoms | Relative Risk ReductionRRR | Absolute Risk ReductionARR | Number Needed to TreatNNT | |
|---|---|---|---|---|
| Omeprazole alone Symptoms”control” Event RateCER | Hp eradication SymptomsExperimental Event RateEER | (CER – EER)/CER | CER – EER | 1/ARR |
| $$93%$$ | $$79%$$ | $$(93% – 79%)/93%= 15%$$ | $$93% – 79%= 14%$$ | $$1/14% = 7pts$$ |
$$
95% text{Confidence Interval (CI) on an}~ mathit{NNT}\
= 1 / (text{limits on the} mathit{CI}~ text{of its ARR})\
= pm1.96 sqrt{frac{mathit{CER}times(1-CER)}{text{# of control pts.}}+ frac{mathit{EER}times(1-EER)}{text{# of exper. pts.}}}\
= 7.3%
$$
Your Calculations
| Relative Risk ReductionRRR | Absolute Risk ReductionARR | Number Needed to TreatNNT | ||
|---|---|---|---|---|
| CER | EER | (CER – EER)/CER | CER – EER | 1/ARR |
| $$93%$$ | $$79%$$ | $$(93% – 79%)/93%= 15%$$ | $$93% – 79%= 14%$$ | $$1/14%= 7pts$$ |
Can you apply this valid, important evidence about a treatment in caring for your patient?
Do these results apply to your patient?
- Is your patient so different from those in the trial that its results can’t help you?
- Yes, she is similar to those patients included in the trial.
- How great would the potential benefit of therapy actually be for your individual patient?
- Method I:$$mathit{f}\\
text{Risk of the outcome in your patient, relative to patients in the trial.
expressed as a decimal}: 1.0\\
mathit{NNT}/mathit{F}\\
= 7/1\\
= 7 (mathit{NNT}~text{for patients like yours})$$ -
Method II: $$1 / (mathit{PEER} times mathit{RRR})\\
text{Your patient’s expected event rate if they received the control treatment:}\\
mathit{PEER}:\_\_\_\_\_\_ \\
1 / (mathit{PEER} times mathit{RRR}) \\
= 1/\_\_\_\_\_\_\_\_ \\
=\_\_\_\_\_\_\_(mathit{NNT}~text{for patients like yours})
$$
Are your patient’s values and preferences satisfied by the regimen and its consequences?
- Do your patient and you have a clear assessment of their values and preferences?
- Needs to be assessed in each patient
- Are they met by this regimen and its consequences?
- Needs to be assessed in each patient
Additional Notes
–
Completed Therapy Worksheet II for Evidence-Based Gastroenterology and Hepatology
Citation
Lack of effect of treating helicobacter pylori infection in patients with nonulcer dyspepsia. Blum A.L. et al N Engl J Med 1998;339:1875-81.
Are the results of this single preventive or therapeutic trial valid?
- Was the assignment of patients to treatments randomised?
- And was the randomisation list concealed?
- Yes
- Were all patients who entered the trial accounted for at its conclusion?
- And were they analysed in the groups to which they were randomised?
- Yes
- Were patients and clinicians kept “blind” to which treatment was being received?
- Yes. The study was placebo controlled.
- Aside from the experimental treatment, were the groups treated equally?
- Yes
- Were the groups similar at the start of the trial?
- No. 66% of the eradication group were men compared to 76% of omeprazole alone group.
Are the valid results of this randomised trial important?
Sample Calculations
| Occurrence of symptoms | Relative Risk Reduction RRR |
Absolute Risk Reduction ARR |
Number Needed to Treat NNT |
|
|---|---|---|---|---|
| Omeprazole alone Symptoms “control” Event Rate CER |
Hp eradication Symptoms Experimental Event Rate EER |
(CER – EER)/CER | CER – EER | 1/ARR |
| $$79.3%$$ | $$72.6%$$ | $$(79.3% – 72.6%)/79.3% = 8.4%$$ |
$$79.3% – 72.6% = 6.7%$$ |
$$1/6.7% = 15text{pts}$$ |
$$95% text{Confidence Interval (CI) on an} ~mathit{NNT}\
= 1 / (text{limits on the CI of its} ~ mathit{ARR})\
= pm 1.96 sqrt{frac{mathit{CER}times(1-mathit{CER})}{text{# of control pts.}} + frac{mathit{EER}times(1-mathit{EER})}{text{# of exper. pts.}}}\
= 9.2%
$$
Your Calculations
| Relative Risk Reduction RRR |
Absolute Risk Reduction ARR |
Number Needed to Treat NNT |
||
|---|---|---|---|---|
| CER | EER | (CER – EER)/CER | CER – EER | 1/ARR |
| $$0.793$$ | $$0.726$$ | $$(0.793 – 0.726)/0.793 = 8.4%$$ |
$$0.067$$ | $$15$$ |
Can you apply this valid, important evidence about a treatment in caring for your patient?
Do these results apply to your patient?
- Is your patient so different from those in the trial that its results can’t help you?
- Yes, she is similar to those patients included in the trial
- How great would the potential benefit of therapy actually be for your individual patient?
- Method I:$$mathit{f}\\
text{Risk of the outcome in your patient, relative to patients in the trial. expressed as a decimal}: 1.0\\
mathit{NNT}/mathit{F}\\
= 15/1\\
= 15 (mathit{NNT}text{for patients like yours})\\
$$ -
Method II:
$$1 / (mathit{PEER}times mathit{RRR})\\
text{Your patient’s expected event rate if they received the control treatment:}\\
mathit{PEER}:\_\_\_\_\_\_ \\
1 / (mathit{PEER}times mathit{RRR})\\
= 1/\_\_\_\_\_\_\_\_
$$
Are your patient’s values and preferences satisfied by the regimen and its consequences?
- Do your patient and you have a clear assessment of their values and preferences?
- Needs to be assessed in each patient.
- Are they met by this regimen and its consequences?
- Needs to be assessed in each patient.
Additional Notes
The eradication group contained a higher proportion of women than the control group. If there is a sex difference in response to treatment the study might have over or underestimated the effect. I might question the validity of this study. Analysis of the data on a “per-protocol” basis provides a higher NNTof 23 due to the large number of exclusions in both randomisation limbs.
Comments on the 2 Studies
- The McColl study had greater validity in that the two groups of randomised patients were better matched and there were no exclusions after randomisation.
- The power of the McColl study was greater due to the larger number of patients treated and analysed.
- The McColl study is more widely applicable in that the definition of dyspepsia was broad and a broad range of severity of symptoms was included.
Nonulcer dyspepsia – Helicobacter Eradication reduces the prevalence of symptoms
Clinical Bottom Line
Helicobacter pylori eradication in patients with nonulcer dyspepsia using omeprazole and two antibiotics for 2 weeks results in symptomatic improvement.
Citation
Symptomatic benefit from eradicating helicobacter pylori infection in patients with nonulcer dyspepsia. McColl K et al. N Engl J Med 1998;339:1869-74.
Clinical Question
In a patient with nonulcer dyspepsia and helicobacter pylori infection, will helicobacter eradication therapy result in a reduction in symptoms of dyspepsia?
Search Terms
“nonulcer dyspepsia” and “helicobacter pylori” and “randomised controlled trial” in MEDLINE
The Study
- Double-blinded concealed randomised placebo controlled trial with intention-to-treat analysis.
- Patients: adults referred with dyspepsia for =4 months, no previous or current peptic ulceration, oesophagitis or NSAID use.
- Control group (N = 160: 160 analysed): placebos
- Experimental group (N = 158; 158 analysed): 20mg omeprazole po; bd, amoxycillin 500mg po;tds, (or tetracycline 500mg po;tds for penecillin sensitivity), metronidazole 400mg po;tds all for 2 weeks VS omeprazole 20mg bd;po + placebo.
The Evidence
| Outcome | Time to Outcome | CER | EER | RRR | ARR | NNT |
|---|---|---|---|---|---|---|
| $$text{Persisting symptoms}$$ | $$text{1 year}$$ | $$0.93$$ | $$0.79$$ | $$15%$$ | $$0.14$$ | $$7$$ |
| $$95% ~text{Condidence Intervals}$$ | $$7.7%text{to}~22.3%$$ | $$4 ~text{to}~ 14$$ |
Comments
- Dyspepsia defined as intermittent or persistent pain or discomfort in upper abdomen or lower chest, heartburn, nausea, postprandial fullness or any other upper GI symptom.
- Carbon-14 urea breath test used to preselect patients with evidence of Helicobacter pylori infection.
- Upper GI endoscopy confirmed diagnosis in all patients
- Analysis of symptoms 1 year after treatment.
