Sodium-glucose cotransporter-2 inhibitors
Use in patients with chronic kidney disease
Jump to comment:
- RE: RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)G. Michael AllanPublished on: 09 January 2023
- RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)Ross A. AveryPublished on: 30 November 2022
- Published on: (9 January 2023)Page navigation anchor for RE: RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)RE: RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)
- G. Michael Allan, MD CCFP, Research (PEER) team
Dr Avery’s letter demonstrates the key principle that leads to most of his grievances: that he wishes for full systematic reviews and guidelines to answer our clinical questions, without concern for word count. For those unfamiliar with TFPs, they are not guidelines (or full systematic reviews) but rather short (350 words maximum) evidence summaries of clinically relevant primary care topics.
Family physicians are overwhelmed with their clinical and administrative work1 as well as the ongoing need to stay up to date in all disciplines of medicine. Information management studies continually point to time as the primary limitation.2-4 TFP addresses this by providing evidence summaries focused on brevity with succinct messages, targeted at busy primary care clinicians.
Many of Dr. Averys points reflect this fundamental difference in the intended definition and purpose of a TFP, thus we will not address each one. As an example of how the individual points could be addressed, we will respond to the last and first points.
The last point is Point 25, which has five subpoints:
1) “The guideline is not supported by evidence,…” TFPs are not guidelines but brief evidence summaries. Then, “not even possible to tell if the NNT cited are statistically significantly different,…” The beginning of the evidence section states “results statistically significant unless noted.”
Show More
2) “The Guideline is overly broad.” It is not clear if Dr Avery is refer...Dr Avery’s letter demonstrates the key principle that leads to most of his grievances: that he wishes for full systematic reviews and guidelines to answer our clinical questions, without concern for word count. For those unfamiliar with TFPs, they are not guidelines (or full systematic reviews) but rather short (350 words maximum) evidence summaries of clinically relevant primary care topics.
Family physicians are overwhelmed with their clinical and administrative work1 as well as the ongoing need to stay up to date in all disciplines of medicine. Information management studies continually point to time as the primary limitation.2-4 TFP addresses this by providing evidence summaries focused on brevity with succinct messages, targeted at busy primary care clinicians.
Many of Dr. Averys points reflect this fundamental difference in the intended definition and purpose of a TFP, thus we will not address each one. As an example of how the individual points could be addressed, we will respond to the last and first points.
The last point is Point 25, which has five subpoints:
1) “The guideline is not supported by evidence,…” TFPs are not guidelines but brief evidence summaries. Then, “not even possible to tell if the NNT cited are statistically significantly different,…” The beginning of the evidence section states “results statistically significant unless noted.”
2) “The Guideline is overly broad.” It is not clear if Dr Avery is referring to the TFP itself, the guideline he mentions, or the guideline cited in the TFP.
3) & 5) “There is no discussion of harms.” And “There is no information on medication costs” These answers are in the original electronic publication5 but unfortunately the points were cut for the journal publication word count.
4) “The guideline should be duration limited,… The data for dapagliflozin extends to 2.4 years.” If we limited recommendations only to the duration of the trials, we would prescribe statins, blood pressure medicines, diabetes agents, etc, for 2-5 years only. This dogmatic adherence to trial duration would transform (negatively) medical practice.Lastly, we take exception to Dr Avery’s point “Tools for Practice appear to be written primarily by pharmacists.” The reference to defend this point is used out of context (we know, we wrote it).6 It refers to advocacy for family physicians on primary care guidelines that have been traditionally dominated by our specialist colleagues. TFP are not guidelines, so the cited reference is out of context. While Family Physicians are involved in all TFPs, the lead author can be any primary care practitioner with an understanding of the evidence and no financial conflicts of interest. The exceptional contribution of pharmacists (and nurses and others) is actively sought and valued. In the age of team-based care, we choose to value the contributions of other health professions. We are fortunate to have broad representation of professional colleagues on this TFP and on our team and celebrate our success in building a team of diverse professionals from around the country. It is disappointing that the profession of authors somehow represents part of an argument to retract an article.
In summary, Dr Avery’s complaints represent a foundational separation of values and misunderstanding. We will not sacrifice the need for succinct, summarized evidence in PEER’s constant efforts to empower primary care through evidence.
[This response is from the core editorial group of TFP. Given the tone and rancor of the letter, we offered to respond for the authors.]
Authors: G. Michael Allan, Christina Korownyk, Michael R Kolber, and Adrienne Lindblad.References
Show Less
1) Porter J, Boyd C, Skandari MR, Laiteerapong N. Revisiting the Time Needed to Provide Adult Primary Care. J Gen Intern Med. 2022 Jul 1. doi: 10.1007/s11606-022-07707-x.
2) Alper BS, Hand JA, Elliott SG, Kinkade S, Hauan MJ, Onion DK, Sklar BM. How much effort is needed to keep up with the literature relevant for primary care? J Med Libr Assoc. 2004 Oct;92(4):429-37.
3) Toulkidis V, Donnelly NJ, Ward JE. Engaging Australian physicians in evidence-based medicine: a representative national survey. Intern Med J. 2005 Jan;35(1):9-17.
4) Cook DA, Sorensen KJ, Wilkinson JM, Berger RA. Barriers and decisions when answering clinical questions at the point of care: a grounded theory study. JAMA Intern Med. 2013 Nov 25;173(21):1962-9.
5) Falk J, Klarenback S, Lam C, Potter J. Should a ‘flozin’ be chosen? Part 2: SGLT2 inhibitors in patients with chronic kidney disease. [Publication on the Internet] Tools for Practice, College of Family Physicians of Canada, 2022 July 12. https://gomainpro.ca/wp-content/uploads/tools-for-practice/1657642591_tf... [Accessed Dec 11, 2022]
6) Allan GM. Should primary care guidelines be written by family physicians? YES. Can Fam Physician Med Fam Can. 2016;62(9):705-706.Competing Interests: None declared. - Published on: (30 November 2022)Page navigation anchor for RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)RE: Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319)
- Ross A. Avery, Family Physician, None
Dear Editor:
Sodium-glucose cotransporter-2 inhibitors:Use in patients with chronic kidney disease AKA TFP #319 has 25 serious errors and problems that I have identified in the evidence-based peer review that follows this introduction. The article Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319) should be retracted. In the past the CFP has based retraction decisions on the COPE guidelines(i). The first reason the guidelines give for retraction is: "Editors should consider retracting a publication 1 if: They have clear evidence that the findings are unreliable, either as a result of major error (eg, miscalculation or experimental error), or as a result of fabrication (eg, of data) or falsification (eg, image manipulation)". This peer review provides clear evidence that TFP 319 has multiple major errors which render the findings unreliable. This means that the criteria for retraction has been easily met.
In addition, the fact that an article with so many errors and problems was published in the CFP journal suggests that the pre-publication review of articles requires improvement.User’s Guide to the peer review: The review starts with a compendium of the identified errors/problems. Each error/problem is then explained in the main review that follows. Structurally, the main review provides (horizontally bordered) extracts of “Tools For Practice” (TFP) #319, followed by a revi...
Show MoreDear Editor:
Sodium-glucose cotransporter-2 inhibitors:Use in patients with chronic kidney disease AKA TFP #319 has 25 serious errors and problems that I have identified in the evidence-based peer review that follows this introduction. The article Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP #319) should be retracted. In the past the CFP has based retraction decisions on the COPE guidelines(i). The first reason the guidelines give for retraction is: "Editors should consider retracting a publication 1 if: They have clear evidence that the findings are unreliable, either as a result of major error (eg, miscalculation or experimental error), or as a result of fabrication (eg, of data) or falsification (eg, image manipulation)". This peer review provides clear evidence that TFP 319 has multiple major errors which render the findings unreliable. This means that the criteria for retraction has been easily met.
In addition, the fact that an article with so many errors and problems was published in the CFP journal suggests that the pre-publication review of articles requires improvement.User’s Guide to the peer review: The review starts with a compendium of the identified errors/problems. Each error/problem is then explained in the main review that follows. Structurally, the main review provides (horizontally bordered) extracts of “Tools For Practice” (TFP) #319, followed by a review of that extract. The references that are intrinsic to the peer review are indicated by bracketed Roman numerals. The references in the TFP extracts have been modified to fit the scheme of this letter.
Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease (TFP 319) : An Evidence Based Peer Review:
A Compendium of the Review Points
General points:
* Tools for Practice appear to be written primarily by pharmacists.
* There is no integration of any the data presented.
* There is no assessment of the quality of any of the data being presented.
*There is no statistical analysis or proper presentation of statistics for any of the results.
*The subtitle is “Use in patients with chronic kidney disease” (CKD) but there is no data presented for CKD patients who do not have diabetes.
*This TFP is promoting off-label prescription medication use.
*Tools for Practice are biased towards the use of pharmaceuticals.Referring to the “Bottom Line”:
*The premise of the “Clinical Question” and the “Bottom Line” statement is not legitimate.
*Patients with different characteristic have been (crudely) combined for the TFP without this being acknowledged.
*The expressions for the treatment effect: “...approximately 3 or 4 fewer” and “3-4 fewer” have no scientific meaning.
*The duration of 5 years is not legitimate.
*There is no inclusion of harms in the “Bottom Line” or anywhere in the published version.
*The Statement that “Sotagliflozin is not better than placebo for these outcomes.” undermines the entire premise of this TFP.Additional points for the first meta-analysis upon which the “Bottom Line” is based:
*The number N=52,827 is misleading.
*There is a numerical error in the NNT.Points referring to the second presented met-analysis:
*The statement “8 RCTs, 26,106 patients” is misleading.
*The NNT for the “Composite Kidney Outcome” does not include all of the relevant data.
*The NNT calculated by the TFP for the “Composite Kidney Outcome” has no duration and therefore has no meaning.Points referring to the three individual studies presented:
*There is no explanation as to why these particular studies are presented.
*There is no explanation as to how these three studies relate to the meta-analyses presented.
*The studies are stated to be industry funded but there is no explanation of what effect this may have on the results.Points referring to the “Context”:
*The “Context” does not provide any context.
*The recommendation reported in the “Context” is not a precise representation of the guideline referenced. *The TFP does not provide any details about the guideline being referred to or its authors or its credibility. *The majority of authors writing the guideline, that is being referred to, have major conflicts of interest.Evidence-Based Peer Review-Main Section
_____________________________________________________________________________________
Sodium-glucose cotransporter-2 inhibitors (ii)
Use in patients with chronic kidney diseaseClinical question
How do sodium-glucose cotransporter-2 inhibitors (SGLT2Is) affect patient outcomes in chronic kidney disease (CKD)?
Bottom line
For every 100 patients with CKD taking an SGLT2I for 5 years, approximately 3 or 4 fewer will develop endstage kidney disease and 3 to 4 fewer will die from any cause versus placebo. Sotagliflozin is not better than placebo for these outcomes.
_____________________________________________________________________________________1) Tools for Practice appear to be written primarily by pharmacists. The first author for this TFP is Jamie Falk, BSc (Pharm) PharmD. By convention the first author listed in scientific articles is the lead author: the one who had the idea or had the most input or did the most work (or any combination of these). The authorship of this TFP is split between two pharmacists, a specialist and a family physician.
The importance of family physician authorship of guidelines directed at family physicians has been reported in the Canadian Family Physician: “Our leadership, the College of Family Physicians of Canada and the provincial chapters, should begin by stating that they will not endorse guidelines targeting primary care unless they are led by primary care physicians and have reasonable and proportional representation from primary care physicians” (iii).2) There is no assessment of the quality of the data presented. The first step in producing any guideline is to assess the quality of the data which underlies the guideline.
3) There is no integration of any the data presented. Data from several meta-analyses and several studies are presented in this TFP, but the “Bottom Line” is derived solely from a single meta-analysis (iv). This is apparent because it is the only study which reports (supposed) 5 year data (see below).
4) There is no statistical analysis or proper presentation of statistics for any of the results. By convention the NNT should be reported with a 95% confidence interval but because there was no statistical analysis in this TFP that did not happen.
NNT were calculated for the purpose of this TFP but it used the statistical significance that was reported in the source scientific article. However, the articles reported the statistics for Odds Ratio (OR) or Hazard Ratio (HR) not for the NNT. The statistics for OR or HR, and NNT are not the same. Their 95% Confidence Intervals (CI) will be different as will their statistical p values. To illustrate this point about the difference: for the same comparison it is possible to have a statistically significant OR or HR and a non-significant NNT (or vice-versa). Calculating statistics for NNT can be complicated(v) but there are online calculators, an example of one is cited (vi).5)The subtitle is “Use in patients with chronic kidney disease” (CKD) but there is no data presented for CKD patients who do not have diabetes. The Bottom Line is based solely on a single meta-anaylsis which has only diabetics as its subjects. Only a single study presented: DPA-CKD includes non-diabetic patients(vii). They make up one third of the subject population but the parameters presented in the TFP are not differentiated with respect to diabetic status. There is no data presented addressing the effects of SGLT2i on non-diabetic patients, so they should not be included in this recommendation.
6) The premise of the “Clinical Question” and the “Bottom Line” statement is not legitimate. The “Bottom Line” statement is derived from the number needed to treat (NNT) but an NNT is for a specific drug at a specific dose. The “Bottom Line” statement is derived from a single meta-analysis that combines data from many drugs within the same class (SGLT2i). The meta-analysis and the “Bottom Line” ignore the fact that different drugs within the same class will have different treatment effects.
Drugs are regulated by Health Canada and the FDA and other national agencies on an individual basis because drugs within a class are not equivalent. For example Vioxx (rofecoxib) and Baycol (cerivastatin) are no longer used, but we are still using other members of their respective classes.
This illegitimate amalgamation of different drugs alone nullifies the “Bottom Line” and indeed the entire TFP #319, however, this assessment will continue so that its other deficiencies may be catalogued.7) This TFP is promoting off-label prescription medication use. Health Canada rules have not been considered by this TGP. Dapagliflozin (Forxiga) is the only SGLT2 inhibitor approved for treatment of all types of CKD in Canada(viii). Canagliflozin (Invokana) is approved for the prevention of progression of diabetic nephropathy(ix). Empagliflozin (Jardiance) does not have an approved role in CKD.
Please note that the SGLT2 inhibitors are not approved for use in patients with Type 1 diabetes due to the risk of diabetic keto-acidosis (DKA). And according to the package information “Forxiga should not be used for the treatment of DKA or in patients with a history of DKA”, those with low eGFR or pediatric patients.8) Patients with different characteristic have been (crudely) combined in this TFP without this being acknowledged. In the meta-analysis used for the “Bottom Line” there were two groups with CKD. One group called high risk who had CKD alone and one group called very high risk who had CKD and cardiovascular disease (CVD). The TFP crudely combined these groups by amalgamating their results and then presenting the result as the outcome for the “CKD group”. But the presence of cardiovascular disease makes a large difference: for the study CKD group the NNT = 39 (2.6 fewer subjects dying per 100 treated for 5 years ) and for the study CKD+CVD group NNT=25 (4 fewer subjects dying per 100 treated for 5 years).
The effect of combining these 2 groups is the formation of an artificial study population where 50% have CKD and 50% have CKD+CVD. So, the results would only apply to a clinical situation where precisely 50% of the CKD patients had cardiovascular disease. In summary, combining groups here creates an artificial group that does not reflect patient populations in clinical situations and also looses information regarding the effect of treatment in those with CKD vs those with CKD and CVD. The TFP should have at least acknowledged that groups were combined and explained why it was done and what it means.9) The expressions for the treatment effect: “...approximately 3 or 4 fewer” and “3-4 fewer” have no scientific meaning. Scientifically “approximately 3 or 4 fewer” has no meaning without further explanation.
The second part of the statement “3 to 4 fewer will die from any cause versus placebo” has a possible explanation. Based on the fact that 2 groups were combined in the “Bottom Line” in the statement the “3” may be for the effect on CKD patients (2.6 fewer subjects per 100 treated for 5 years dying) and the “4” may be for the effect on CKD+CVD patients (4.0 fewer subjects per 100 treated for 5 years dying) . Regardless, the expressions “...~3 or 4 fewer...” and “...3-4 fewer...” are both unexplained and inappropriate.
In addition, by convention NNT are always rounded up, which is not followed in this TFP.10) The duration of 5 years is not legitimate. The meta-analysis used for the “Bottom Line” reports five-year data but the included studies did not last that long, meaning the data was extrapolated. There is no basis in evidence to extrapolate the results of studies beyond the actual duration of the studies.
11) There is no inclusion of harms in the “Bottom Line” or anywhere in the published version of this TFP. The information on adverse events that appeared in the version distributed by the CFPC has been removed from the published version. The “Bottom Line” is being used as a form of mini-guideline for a clinical decision. Inclusion of harms in a guideline is necessary so that benefits can be weighed against harms when making a clinical decision.
The harms to be considered for these drugs include: mycotic genital area skin/mucosal infections, diabetic keto-acidosis, volume depletion, and the possibility of increased lower limb amputations with canagliflozin(x). With SGLT2i recognizing harms is even more important, as strategies have been suggested that may decrease the incidence of the harms(x).
It was the awareness of harmful effects that led to the discontinuation of Fen-Phen, Propulsid, Baycol, Vioxx, etc.12) The Statement that “Sotagliflozin is not better than placebo for these outcomes.” undermines the entire premise of this TFP. The claim that is being made through the “Bottom Line” is that treatment with a SGLT2i will result in 3-4 fewer deaths for 100 patients treated over 5 years compared with a placebo. But sotagliflozin is a SGLT2i and it does not have this effect. Which means that not every SGLT2i has the same effect on all cause mortality and that these drugs should be treated individually and not together as a class when looking at this outcome (this applies to the other outcomes discussed in this TFP).
The idea that these drugs should be analysed individually was explained above and the TFP has provided a specific example as to why this is so._____________________________________________________________________________________
EvidenceResults were statistically significant unless noted.
• Two systematic reviews of relevant RCTs included
patients with CKD.(iv,xi)-In 1 review of 52,827 patients with cardiovascular
and CKD risks, among those with CKD at 5 years(iv):
—End-stage kidney disease: 8.9% vs 12% (placebo),
number needed to treat (NNT)=33.
—Cardiovascular death: 11% vs 14% (placebo), NNT=27.
—Overall mortality: 19% vs 22% (placebo), NNT=31
_____________________________________________________________________________________This is the first meta-analysis presented, it is the one that provides all the data for the “Bottom Line” and is critiqued above. As mentioned before its problems include: combining many different drugs as well as multiple doses for some drugs into a single meta-analysis. And also, the results were extrapolated to 5 years which has no justification.
13) The number N=52,827 is misleading. This is the total number of subjects in the SGLT2i studies but since not all of them had CKD the applicable N is 8,231. I calculated it by adding all the studies marked as “chronic renal disease” in the data supplement (xii).
14) There is a numerical error in the NNT. The TFP calculated overall mortality should be 18% vs 22% and not 19 vs 22%.
_____________________________________________________________________________________
-In the other review (8 RCTs) involving 26,106 patients with baseline CKD, at 2.5 years(xi):—Cardiovascular disease: 10% vs 11% (placebo), NNT=91.
—Composite kidney outcome (40% to 60% estimated glomerular filtration rate decline, end-stage kidney disease, or renal death): 4.8% vs 6.9% (placebo), NNT=48.
-Limitations: included RCTs not specific to CKD patients
_____________________________________________________________________________________This is the Second meta-analysis that is presented. The composite Kidney Outcome is used as an example here, similar problems apply to the outcome called Cardiovascular Disease.
15) The statement “8 RCTs, 26,106 patients” is misleading because the TFP uses only 4 RCTs (randomized controlled trial) and 19,156 “patients” in their calculation of the “Composite Kidney Outcome”.
16) The NNT for the “Composite Kidney Outcome” does not include all of the relevant data. This is another calculated NNT for the purpose of this TFP; calculated from Fig. 1 Table B in the reference. Because one of the studies does not include the data needed to calculate NNT in the table, it is left out of the NNT calculation without informing the reader.
17) The NNT calculated by the TFP for the “Composite Kidney Outcome” has no duration and therefore has no meaning. A NNT is always for a given duration. NNT was calculated for this TFP using the sum of the proportion of the cases in the placebo vs that of the SGLT2i groups from the 4 studies from Fig 1B in the reference. The problem is that the studies have different durations: 14.2-50.4 months; so the NNT produced has no corresponding duration and is therefore meaningless.
_____________________________________________________________________________________
• In 3 industry-funded RCTs (with CKD patients),3-5 mean
estimated glomerular filtration rate was about 40 to 55 mL/
min/1.73 m2, albumin-to-creatinine ratio was about 75 to
105 mg/mmol, and 67% to 100% of patients had diabetes.-In a trial of 4401 patients taking 100 mg of canagliflozin daily(xiii), at 2.6 years:
—End-stage kidney disease: 5.3% vs 7.5% (placebo),
NNT=45.
—Cardiovascular death: 5.0% vs 6.4% (placebo), NNT=71.
—All-cause mortality: 7.6% vs 9.1% (placebo), NNT=67.-In another trial of 4304 patients taking 10 mg of
dapagliflozin,(vii) at 2.4 years:
—End-stage kidney disease: 5.1% vs 7.5% (placebo),
NNT=42.
—Cardiovascular death: 3.0% vs 3.7% (placebo), not
statistically different.
—All-cause mortality: 4.7% vs 6.8% (placebo), NNT=48-In a third trial of 10,584 patients taking 200 mg to
400 mg of sotagliflozin daily,(xiv) at 1.3 years:
—No differences in composite kidney outcome,
cardiovascular death, or all-cause mortality.
__________________________________________________________________________________________________________________________18) There is no explanation as to why these particular studies are presented. There are many studies to choose from, for example the first meta-analysis cites 14 studies relevant to this clinical question. I presume that it is because these are the studies that were carried out by the pharmaceutical companies with a view to meeting the requirements for regulatory approval or revision.
19) There is no explanation as to how these three studies relate to the meta-analyses presented.
20) The studies are stated to be industry funded but there is no explanation of what effect this may have on the results. TFP often indicate that studies are industry sponsored but do not explain its implications. From a Cochrane review of industry sponsorship “Industry sponsored studies more often had favorable efficacy results, RR: 1.27 (95% CI: 1.17 to 1.37) (25 papers) (moderate quality evidence)”. These findings of more favourable efficacy could not be explained by a standard risk of bias assessment.(xv)
__________________________________________________________
Context
• Metformin and SGLT2Is are recommended first-line therapies for those with type 2 diabetes and CKD.(xvi)
__________________________________________________________21) The recommendation reported in the “Context” is not a precise representation of the guideline referenced. The KDIGO guideline contains a restriction based on estimated glomarular filtration rate (eGFR) :”We recommend treating patients with T2D, CKD, and an eGFR ≥30 ml/min per 1.73 m2 with metformin (1B). We recommend treating patients with T2D, CKD, and an eGFR ≥30 ml/min per 1.73 m2 with an SGLT2i (1A)”(xvi).
22) The TFP does not provide any details about the guideline being referred to or its authors or its credibility. A better known guideline with somewhat fewer conflicts of interest would be: Diabetes Canada, their guidelines suggest a similar course of action:
In adults with type 2 diabetes and CKD and an estimated eGFR >30 mL/min/1.73m2: An SGLT2i should be used to reduce the risk of:
Progression of nephropathy [Grade A, Level 1A (16) for canagliflozin; Grade A, Level 1 (18) for empagliflozin and dapagliflozin].
HHF [Grade A, Level 1 (18) for canagliflozin, dapagliflozin and empagliflozin].
MACE [Grade B, Level 2 for canagliflozin (16), Grade C, Level 3 (12) for empagliflozin].
A GLP1-RA may be considered to reduce the risk of MACE (Grade B, Level 2 (6,7) for liraglutide and semaglutide)(xvii).
Interestingly, Diabetes Canada does not adhere to Health Canada Guidelines.23) The majority of authors writing the guideline, that is being referred to, have major conflicts of interest. 13 of 24 authors declared a conflict of interest including 2 authors who are on the board of two pharmaceutical manufacturers of SGLT2i. Interestingly, the importance of minimizing the number of guideline authors with conflicts of interest has been reported in the Canadian Family Physician(xxiv).
24) The “Context” does not provide any context. The type of information that could be provided here would include 1) The magnitude of the problem of CKD. 2) Any differences to the approach to CKD in diabetic vs non-diabetic patients, 3) The biological mechanism by which an SGLT2i is supposed to positively affect CKD (is it related to blood sugar control or it just an expensive diuretic?), 4) the appropriateness of using NNT to describe the effects of a preventative treatment when average number of months of added longevity is likely a more appropriate measure.
25) Tools for Practice are biased towards the use of pharmaceuticals.
Most of these arguments have already been explained earlier in this review so are presented succinctly here.
1) Their guideline is not supported with evidence. In TFP #319 the only evidence in the “Bottom Line” is a single meta-analysis of poor quality. And it is not even possible to tell if the NNT cited are statistically significantly different because there is no statistical analysis for NNT.
2) The guideline is overly broad. TFP #319 does not consider Health Canada Guidelines that permit only dapagliflozin for all types of CKD (with restrictions as already noted), and they have included non-diabetics in their guideline without providing any supporting evidence for this inclusion.
3) There is no discussion of harms- a guideline must always be a balance of benefits and harms.
4) The guideline should be duration limited to reflect the duration of the evidence. The data for dapagliflozin extends to 2.4 years, so the guideline for this drug should limit its recommended use in CKD to 2.4 years.
5) There is no information on medication costs. The cost of medication may be an important factor in clinical decisions.References
i.Barbour V, Kleinert S, Wager E, Yentis S. Guidelines for Retracting Articles. Committee on Publication Ethics; 2009. doi:10.24318/cope.2019.1.4
ii.Falk J, Klarenbach S, Lam C, Potter J. Sodium-glucose cotransporter-2 inhibitors: Use in patients with chronic kidney disease. Can Fam Physician. 2022;68(10):753-753. doi:10.46747/cfp.6810753
iii.Michael Allan G. Should primary care guidelines be written by family physicians? YES. Can Fam Physician Med Fam Can. 2016;62(9):705-706.
iv.Palmer SC, Tendal B, Mustafa RA, et al. Sodium-glucose cotransporter protein-2 (SGLT-2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists for type 2 diabetes: systematic review and network meta-analysis of randomised controlled trials. BMJ. Published online January 13, 2021:m4573. doi:10.1136/bmj.m4573
v.Altman DG. Confidence intervals for the number needed to treat. BMJ. 1998;317(7168):1309-1312. doi:10.1136/bmj.317.7168.1309
vi.Computing number needed to treat (NNT). Accessed October 30, 2022. https://www.graphpad.com/quickcalcs/NNT1.cfm
vii.Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al. Dapagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2020;383(15):1436-1446. doi:10.1056/NEJMoa2024816
viii.AstraZeneca. Forxiga Product Monograph. Published online August 6, 2021. https://www.astrazeneca.ca/content/dam/az-ca/downloads/productinformatio...
ix.Johnson JPC of J&. Health Canada Approves New Indication for INVOKANA®* (canagliflozin) to Reduce the Risks Associated with Diabetic Kidney Disease in Patients with Type 2 Diabetes. Accessed October 23, 2022. https://www.newswire.ca/news-releases/health-canada-approves-new-indicat...
x.Milder TY, Stocker SL, Day RO, Greenfield JR. Potential Safety Issues with Use of Sodium-Glucose Cotransporter 2 Inhibitors, Particularly in People with Type 2 Diabetes and Chronic Kidney Disease. Drug Saf. 2020;43(12):1211-1221. doi:10.1007/s40264-020-01010-6
xi.Kaze AD, Zhuo M, Kim SC, Patorno E, Paik JM. Association of SGLT2 inhibitors with cardiovascular, kidney, and safety outcomes among patients with diabetic kidney disease: a meta-analysis. Cardiovasc Diabetol. 2022;21(1):47. doi:10.1186/s12933-022-01476-x
xii.Supplementary Data: Sodium-glucose cotransporter protein-2 (SGLT-2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists for type 2 diabetes: systematic review and network meta-analysis of randomised controlled trials. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7804890/bin/pals058523.ww.pdf
xiii.Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy. N Engl J Med. 2019;380(24):2295-2306. doi:10.1056/NEJMoa1811744
xiv.Bhatt DL, Szarek M, Pitt B, et al. Sotagliflozin in Patients with Diabetes and Chronic Kidney Disease. N Engl J Med. 2021;384(2):129-139. doi:10.1056/NEJMoa2030186
xv.Lundh A, Lexchin J, Mintzes B, Schroll JB, Bero L. Industry sponsorship and research outcome. Cochrane Database Syst Rev. 2017;2:MR000033. doi:10.1002/14651858.MR000033.pub3
xvi.de Boer IH, Caramori ML, Chan JCN, et al. Executive summary of the 2020 KDIGO Diabetes Management in CKD Guideline: evidence-based advances in monitoring and treatment. Kidney Int. 2020;98(4):839-848. doi:10.1016/j.kint.2020.06.024
xvii.Diabetes Canada | Clinical Practice Guidelines - Chapter 13: Pharmacologic Glycemic Management of Type 2 Diabetes in Adults: 2020 Update. Accessed October 21, 2022. https://guidelines.diabetes.ca/cpg/chapter-13-2020-update
Show LessCompeting Interests: None declared.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.