See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/261371351
Response to Lankarani Article in The American Journal of Gastroenterology · April 2014 DOI: 10.1038/ajg.2014.16 · Source: PubMed
CITATIONS
READS
0
105
4 authors: Masoud Mohammad Malekzadeh
Hooman Khademi
Digestive Diseases Research Institute
International Agency for Research on Cancer
8 PUBLICATIONS 41 CITATIONS
47 PUBLICATIONS 719 CITATIONS
SEE PROFILE
SEE PROFILE
Reza Malekzadeh
Farin Kamangar
Tehran University of Medical Sciences
Morgan State University
702 PUBLICATIONS 18,139 CITATIONS
182 PUBLICATIONS 7,523 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Prospectibve Epidemiological Research Studies of Iranian Adults (PERSIAN) Cohort View project
Pancreatobiliary disorders View project
All content following this page was uploaded by Reza Malekzadeh on 09 April 2014. The user has requested enhancement of the downloaded file.
600
Letters to the Editor
We acknowledge that our study population had limited comorbidities owing to the use of a low-volume purgative (sodium picosulfate and maganesium citrate); we excluded patients with volume issues (ascites, congestive heart failure, renal insufficiency) and those who could not give consent. Approximately 25% of patients in each group were on medications one could consider constipating. Although we agree that our study population was typical of a mid-sized Canadian city, we are puzzled by some of the other critiques. Hassan et al. (1) list several factors associated with poor prep, and we have no issue with any of those. However, whether having a breakfast would adversely affect the quality of the prep even in these “difficult” to prep patients certainly has not been shown. In fact, given our results, one could postulate that the opposite may be true—tolerance of the prep could be better and thus compliance may be higher. Nonetheless, we certainly agree that a direction that is yet to be adequately explored to date is how to obtain good bowel cleansing in patients who have predictors of poor preparation or have had a poor prep in the past. CONFLICT OF INTEREST The authors declare no conflict of interest. REFERENCES 1. Hassan T, Periyalwar P, Tahir S et al. Low-residue diet with bowel prep: not for everyone! World J Gastrointest Endosc 2014;109:599 (this issue). 2. Melicharkova A, Flemming J, Vanner S et al. A low-residue breakfast improves patient tolerance without impacting quality of low-volume colon cleansing prior to colonoscopy: a randomized trial. Am J Gastroenterol 2013;108:1551–5. Gastrointestinal Diseases Research Unit, Hotel Dieu Hospital, Queen’s University, Kingston, Ontario, Canada. Correspondence: Lawrence Hookey, MD, Gastrointestinal Diseases Research Unit, Hotel Dieu Hospital, Queen’s University, 166 Brock Street, Kingston, Ontario, Canada K7L 5G2. E-mail:
[email protected]
Opium, an Oncogene or a Marker of Oncogenic Behaviors? Kamran B. Lankarani, MD1 doi:10.1038/ajg.2014.15
The American Journal of GASTROENTEROLOGY
To the Editor: For decades, opium use has been a major health problem in countries such as Iran (1). Opium has wmany adverse effects on health and its association with increased all-cause mortality has been reported in the past, but a causal role for opium in oncogenesis has been a subject of debate. A recent report by Malekzadeh et al. (2) on opium use and risk of mortality from digestive diseases claimed this causality especially for esophageal cancer. Even in large prospective cohort studies like this excellent research with a standard scientific design, causality cannot be concluded. Instead, what could be emphasized are still its associations. Their own data, especially on former opium users, which are presented in Table 2 of Malekzadeh et al. (2), showed that an increased risk of gastrointestinal mortality is not seen in this subgroup, as none of the confidence intervals for them were statistically significant. The exclusion of first-year mortality from their analysis, as shown in Table 3, decreased the power of association and moved the hazard ratio toward the null, although it was still statistically significant. One wonders what would be the result if the exclusion period were extended. Regarding these points, the possibility of reverse causality should still be taken into consideration. Although the authors have adjusted the results for known confounding factors such as age, sex, and cigarette smoking, other confounders cannot be excluded. For instance, the association between drinking hot tea and esophageal squamous cell carcinoma has been reported from this cohort study in the past (3). It is quite likely that opium users, whose pain threshold has increased, do not sense the pain of drinking hot beverages and expose their esophagus to more thermal injury. The association between opium use and vitamin deficiencies is another possibility that was recently reported by this cohort study (4). The increased mortality from cirrhosis in opium users is also of interest. This might be related to more encephalopathy or more infections, especially pneumonia, in opium users with cirrhosis, not necessarily indicating causality. Some of the findings in this research are quite amazing. For instance, an increased malignant as well as nonmalignant gastro-
intestinal death rate in those with higher education is of concern. Confounding factors such as smoking may have had a role as well. In conclusion, this interesting report still cannot draw any conclusions on the causality of opium for esophageal cancer and cirrhosis. CONFLICT OF INTEREST
The author declares no conflict of interest. REFERENCES 1. Lankarani KB, Alavian SM, Peymani P. Health in the Islamic Republic of Iran, challenges and progresses. Med J Islamic Repub Iran 2013;27:42–9. 2. Malekzadeh MM, Khademi H, Pourshams A et al. Opium use and risk of mortality from digestive diseases: a prospective cohort study. Am J Gastroenterol 2013;108:1757–65. 3. Islami F, Pourshams A, Nasrollahzadeh D et al. Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ 2009;338:b929. 4. Malekshah AF, Kimiagar M, Pourshams A et al. Vitamin deficiency in Golestan Province, northern Iran: a high-risk area for esophageal cancer. Arch Iran Med 2010;13:391–4. 1
Health Policy Research Center, Shiraz University of Medical Sciences, Shiraz, Iran. Correspondence: Kamran B. Lankarani, MD, Health Policy Research Center, Shiraz School of Medicine, Zand Blvd., 71348, Shiraz, Iran. E-mail:
[email protected]
Response to Lankarani Mohammad Masoud Malekzadeh, MD1, Hooman Khademi, MD, MPH1,2, Reza Malekzadeh, MD1 and Farin Kamangar, MD, PhD1,3 doi:10.1038/ajg.2014.16
To the Editor: We fully agree with the comment of Lankarani (1) on our paper (2) that it is premature to assertively state that opium causes cancer. There are alternative explanations, such as confounding, reverse causality, biases, and random error. However, our paper is only one among a total of 18 reports that have shown an association between opium use and higher risk of cancer. These reports, which have been summarized in a recent systematic review (3), nearly unanimously show an association VOLUME 109 | APRIL 2014 www.amjgastro.com
Letters to the Editor
between opium use and increased risk for various cancers, including cancers of the larynx, lung, esophagus, stomach, and bladder. It is unlikely that all of these studies could have found these associations solely due to random error. These studies have found relatively strong associations between opium use and cancer, stronger than tobacco use, which make the argument for confounding less likely. The majority of these studies have adjusted or matched for major confounders, including age, sex, and tobacco use. Assuming that high-temperature tea consumption is a confounder for the association between opium use and esophageal cancer, it cannot be a confounder for other types of cancer, such as bladder cancer. The association between opium use and risk for various cancers remained statistically significant after various sensitivity analyses in cohort studies (4). Furthermore, laboratory studies have shown that it is biologically plausible that opium could cause cancer (5). Again, although we agree with Lankarani (1), and suggest that further studies be conducted in this field, the level of evidence accumulated thus far in favor of carcinogenicity of opium is not trivial. We also agree that the association seen between opium use and liver cirrhosis may not be causal. The same alternative explanations apply here too. However, we are obliged to report the associations and offer a balanced discussion in favor or against causality, which we have done so to the best of our knowledge and ability. Further epidemiologic, animal, and laboratory analyses will shed light on whether the association is truly causal. The association between educational level and mortality from malignant and nonmalignant gastrointestinal disorders, shown in Table 1, (2) is based on very small numbers. Hence, that association may be entirely due to random error. However, data on education and gastrointestinal mortality were provided solely to examine potential confounders; education was not a main exposure in our report and further evaluations of the relationship between education and mortality in the Golestan Cohort requires a comprehensive analysis. CONFLICT OF INTEREST
The authors declare no conflict of interest. © 2014 by the American College of Gastroenterology View publication stats
REFERENCES 1. Lankarani KB. Opium, an oncogene or a marker of oncogenic behavior? Am J Gastroenterol 2014;109:600 (this issue). 2. Malekzadeh MM, Khademi H, Pourshams A et al. Opium use and risk of mortality from digestive diseases: a prospective cohort study. Am J Gastroenterol 2013;108:1757–65. 3. Kamangar F, Shakeri R, Malekzadeh R, Islami F. Opium use: an emerging risk factor for multiple cancers? Lancet Oncol 2014;15:e69–e77. 4. Khademi H, Malekzadeh R, Pourshams A et al. Opium use and mortality in golestan cohort study: prospective cohort study of 50 000 adults in Iran. BMJ 2012;344:E2502. 5. Friesen M, O’neill IK, Malaveille C et al. Characterization and identification of 6 mutagens in opium pyrolysates implicated in oesophageal cancer in Iran. Mutat Res 1985;150:177–91. 1
Digestive Disease Oncology Center, Digestive Disease Research Institute, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran; 2International Agency for Research on Cancer, Lyon, France; 3Department of Public Health Analysis, School of Community Health and Policy, Morgan State University, Baltimore, Maryland, USA. Correspondence: Farin Kamangar, MD, PhD, Department of Public Health Analysis, School of Community Health and Policy, Morgan State University, Portage Avenue Campus, Room 302, Baltimore, MD 21251, USA. E-mail:
[email protected]
Re: Proton Pump Inhibitors and Risk for Recurrent Clostridium difficile Infection Among Inpatients Jason Abdallah, MD1, Tuyyab Hassan, MD1 and Annette Kyprianou, MD1 doi:10.1038/ajg.2013.481
To the Editor: We read with interest the article by Freedberg et al. (1) regarding proton pump inhibitors (PPIs) and risk for recurrent Clostridium difficile infection (CDI) among inpatients. The article does not provide adequate information regarding the number, class, or duration of antibiotics used, which as we describe can significantly affect the risk of CDI. Several studies (2,3) describe the association between the use of multiple antimicrobials and an increased risk of CDI.
For example, a seminal case–control study (2) found that patients who developed CDI were more likely than matched controls to have received multiple antimicrobials (80% vs. 56%; P < 0.002). A retrospective cohort study (3) found that the incidence of CDI increased with the number of antimicrobials administered (relative risk, 2.01; 95% confidence interval, 1.67–2.40). Prolonged antimicrobial therapy has also been associated with an increased risk of CDI. A retrospective cohort study of 293 hospitalized patients with CDI found an association between the extended use of antimicrobial therapy and an increased risk of CDI, even after adjustment for other risk factors (4). The effect different antibiotics have on the risk for CDI also needs to be addressed. Several studies (5–7) have been conducted to demonstrate the relationship between different classes of antimicrobials and the risk of both hospital- and communityassociated CDI (CA-CDI). One published meta-analysis (5) considered the impact of antibiotic exposure on CDI risk among hospital inpatients. The meta-analytic study noted that tetracyclines and penicillins were associated with the lowest risk, whereas fluoroquinolones, clindamycin, and expanded-spectrum cephalosporins were associated with the highest risk of CDI acquisition. Another meta-analysis (6) evaluated the risk of CA-CDI from different antibiotics. Their results suggest that certain antibiotics commonly used in outpatient settings are associated with an increased risk of CA-CDI. Clindamycin, fluoroquinolones, and cephalosporins were frequently associated with CA-CDI, whereas penicillins, macrolides, and sulfonamides/trimethoprim were implicated to a lesser extent. Notably, tetracyclines were not associated with an increased risk for CA-CDI. These observations are important, given the possibility of patients receiving antibiotics before being admitted to the hospital. It was mentioned in the article that over 98% of the patients receiving PPIs were given esomeprazole. The relationship between the type of PPI and recurrent CDI is rather under-investigated but it should be noted that when compared with other PPIs (lansoprazole, omeprazole, pantoprazole, The American Journal of GASTROENTEROLOGY
601
