Hyponatraemia reversibly affects human myometrial contractility. An in vitro pilot study

Background In a previous study we found a significant correlation between dystocia and hyponatraemia that developed during labour. The present study examined a possible causal relationship. In vitro studies often use area under the curve (AUC) determined by frequency and force of contractions as a measure of myometrial contractility. However, a phase portrait plot of isometric contraction, obtained by plotting the first derivate of contraction against force of contraction, could indicate that bi-or multiphasic contractions might be less effective compared to the smooth contractions. Material and methods Myometrial biopsies were obtained from 17 women undergoing elective caesarean section at term. Each biopsy was divided into 8 strips and mounted isometrically in a force transducer. Seven biopsies were used in the first part of the study when half of the strips were immersed in the hyponatraemic study solution S containing Na+ 120 mmol/L and observed for 1 hour, followed by 1 hour in normonatraemic control solution C containing Na+ 136 mmol/L, then again in S for 1 hour, and finally 1 hour in C. The other half of the strips were studied in reverse order, C-S-C-S. The remaining ten biopsies were included in the second part of the study. Response to increasing doses of oxytocin (OT) in solutions S and C was studied. In the first part of the study we calculated AUC, and created phase portrait plots of two different contractions from the same strip, one smooth and one biphasic. In both parts of the study we registered frequency and force of contractions, and described appearance of the contractions. Results First part of the study: Mean (median) contractions per hour in C: 8.7 (7.6), in S 14,3 (13). Mean (SD) difference between groups 5.6 (4.2), p = 0.018. Force of contractions in C: 11.8 (10.2) mN, in S: 10.8 (9.2) mN, p = 0.09, AUC increased in S; p = 0.018. Bi-/multiphasic contractions increased from 8% in C to 18% in S, p = 0.001. All changes were reversible in C. Second part of the study: Frequency after OT 1.65 x 10−9 M in C:3.4 (2.9), in S: 3.8 (3.2), difference between groups: p = 0.48. After OT 1.65 x 10−7 M in C: 7.8 (8.9), increase from previous OT administration: p = 0.09, in S: 8.7 (9.0), p = 0.04, difference between groups, p = 0.32. Only at the highest dose of OT dose was there an increase in force of contraction in S, p = 0.05, difference between groups, p = 0.33. Initial response to OT was more frequently bi/multiphasic in S, reaching significance at the highest dose of OT(1.65 x 10−7 M), p = 0.015. when almost all contractions were bi/multiphasic. Conclusion Hyponatraemia reversibly increased frequency of contractions and appearance of bi-or multiphasic contractions, that could reduce myometrial contractility. This could explain the correlation of hyponatraemia and instrumental delivery previously observed. Contractions in the hyponatraemic solution more frequently showed initial multiphasic contractions when OT was added in increasing doses. Longer lasting labours carry the risk both of hyponatraemia and OT administration, and their negative interaction could be significant. Further studies should address this possibility.

Prior paper 2009 looked at oxytocin in multvar regression, but using 5 U total dose and bimodal analysis, showed .072 p value, although CI -.9 -7.4. Thus, probably should not strongly say that oxytocin is not related to hyponatremia, as noted in this abstract. Not to review a 2009 publication, but excessive oxytocin may cause low Na in some patients.
Also no mention of inappropriate ADH due to labor in either paper Phase plot reference 8 should be 10, page 14. The phase plot analysis proposed by the Warwick group actually was not intended primarily as a measure of effectiveness of contractions (especially in vivo), rather normalizing results among different samples.
While customary in the field of uterine contractility, the area under the curve (AUC) is actually a forcetime integration, which is a mechanical impulse. This unfortunately was also chosen decades ago as the word to describe an action potential, hence the potential (pun intended) confusion in a field where the tissue emits both an impulse in the form of an action potential and an impulse in the form of an integrated force. Just a comment, not a recommendation for change, since that boat left dock a while ago. If would be nice to occasionally see the correct terms used, however.

Intro -
• Should mention oxytocin at least as a possibility for causing hyponatremia • Cl-is largely distributed by passive electrical propertiesdonnan/Nernst • Intro describing the action potential is a bit long and the main point is only that these ions create electrical activity that "opens" voltage gated calcium channels which initiates the contraction. If you want to go into detail, you should also mention the Na/Ca exchanger which is probably more important in Na metabolism than the Na channel, esp regarding duration of contractions Sampling • 4 years to write manuscript?
• Omit humanitarian or define it. Given the routine nature of this biopsy, we really don't even need to know other than "clinical indications" and implying that none were done for research purposes alone. • Typo labouratory -"were" to where • Putting normonatraemic Tyrode's once (earlier) is enough, redundant is redundant.
Tyrode's is normal Na. Also, you realize that Mg2+ is a potential tocolytic, probably should comment at some point. Most groups avoid Mg, although for these studies it probably makes no difference. For storage/transport on ice, you should mention pH, though since real Tyrode's is ~ 6.5 unless outgassed with CO2 • Did you stabilized the 100 mg (9.8 mN) tension after tissue creep (i.e. rest period)?
• You studied one biopsy at a time, although up to 8 strips simultaneously, correct?
First part of the study • No need to repeat studying one biopsy at a time • Phase plot reference is Gullum -10. My guess is that you need to use endnote or similar program, since it looks like several of your references are incorrectly numbered Second part of the study • No need to repeat studying one biopsy at a time • This description of oxytocin concentrations brings into question if the 82.5 nM oxytocin was used

Solutions
• While this is a complete description, it would be good to not repeat all the unchanged chemicals, but simply state the that the S solution was made hyponatremic and hypoosmotic by reduction of NaCl to 97.4 mM from 114 mM. Also, if oxytocin was added to the rest period (which I am not sure), you could place oxytocin here as well so that it is clear all tissues were exposed to oxytocin. On the other hand, if all tissues were not exposed to oxytocin, please correct your description on page 7 (top)

Statistics
• Do you mean "mean values for each biopsy..?or each tissue strip?

Results
• Exclusion due to technical problems, or simply the tissue strip failed to contract (which can be up to 20% and still be OK) • Now you have to define humanitarian. I think you mean "elective" • Earlier you probably should note that none of your subjects were in spontaneous labor

First part
• First sentence is not a sentence • Needs to be really clearpeak force is the force change from "resting" to maximum force generated. (or however you measured) • Fig 1 is confusing. The 1

h is not clear. How many strips is this? Upper panel is noted,
but not lower panel. Needs much better labeling • Fig 2. As above, amplitude is probably peak force, should note if measured from onset to peak or simply peak from absolute 0. Second part • This is confusing to readcould this just be a table?
• All the figure legends should be put together at the end, near the figures.

Discussion
• This is where the concept of peak force rather than just force helps keep confusion with AUC (which to some is a force) low. • All changes were reversible in the…. Indicating that hyponatremia or hypoosmolarity … • Probably could mention Parkington's work, but it is a bit overboard to attribute your results to spike-like rather than plateau potentials. • The paragraph that starts with tachysystole doesn't hang together well. You never really got close to 5 contractions in 10 min, so I would suggest you avoid this topic • You really short-changed the discussion on the 2 nd part of the study. What do you think caused multiphasic appearance? What is the physiological underpinning, and why would OT bring that out? • Pulsatile OT is neither here nor there -suggest you either justify or omit • What about multiphasic contractions in vivodo your multiphasic relate to them (doubling, tripling, etc) • Given the effort on phase plotsthey are also a way to quantify multiphasic behaviorthat is the areas of the closed circles of the plots return a measure of the impulse attributable to each phase of the contraction. I actually don't think this must be done because it is a lot of calculating for very little information, but just to point out that it could be done for the 2 nd part oxytocin effects.
Streghts (sic-type) and limitations • Our studies on excised human tissue may more closely represent the physiology of human labor better than rodent model. Or similar phrasing • OT was used to initiate contractions and establish a constant starting point to investigate the specific effects of hyponatremia/ hypoosmolarity on uterine contractions. • You already mention type II error, no need to have the last two lines as a limitation • The key limitation was that you studied hyponatremia without maintaining osmolarity constant, so the effects of Na are mixed with hypoosmolarity. However, since this is the clinical condition you wished to mimic, that may be what you intended to do.

Conclusion
• Some of the other factors for dystocia could be …. For example Sue Wray's group showed that pH is a key factor as well. • Fluid intake was a conclusion of your prior paper, so you probably should merely say that low Na is a modifiable clinical parameter In summary This is a very good paper with very good data. While the philosophy of PLoS One is to publish data regardless of perceived relevance, this paper has both good data and relevance. As pointed out above, there are a few areas of potential confusion and a few missed opportunities. The authors should consider most of this review as suggestions, with questions requiring answers are written in bold.
This manuscript satisfies PLoS One criteria for publication