Hx & Use of Auscultation to replace routine c-EFM in healthy labor patients with normal pregnancies ~ Part 3 of 3

by faithgibson on March 11, 2019

in Cesarean Politics, OB Interventions: Dubious or Detrimental

 Part 3 ~ 2019 

Easy shortcut to share –> http://tinyurl.com/y65wzon5        word count 4700

How did the extreme discordance btw the scientific studies and current obstetrical standard of care come about?

A specially-designed placenta-percreta operating suite outfitted with the latest interventional radiological equipment for performing Cesarean-hysterectomy on previous C-section mothers. In addition to multi-million dollar equipment, there are over 20 doctors, OR nurses, neonatal nurses and specialty hospital staff.



A quick history of fetal monitoring: Intermittent Auscultation & EFM ~ verbs, nouns, and names of physician-inventors

 

The first person to report hearing the sound of an unborn baby’s heartbeat was a 17th century physician by the name of Dr. Marsac. A contemporary of his, a Dr. Killian, was the first to ponder the possibility that the rate and rhythm of the fetal heart might be an indicator of how well the fetus was (or wasn’t) tolerating the process of childbirth.

However, this possibility went unnoticed for two more centuries, until 1818, when doctors Mayor and Kergaradec described a method of  auscultating fetal heart sounds by placing one’s ear on or very near the maternal abdomen. Dr. Kergaradec also suggested that fetal heart sounds could be used to determine fetal viability and proof of life in the months before ‘quickening’, that is, before the mother could feel the baby kick.

In 1833, an English physician, Dr. Evory Kennedy, published guidelines for fetal distress and also recommend auscultation of the fetal heart rate to monitor the well-being of the fetus during labor and birth.

A simple tool for counting the fetal heart rate

However, the big leap forward came in 1893, when Dr. Von Winkel established criteria for fetal distress. After listening to the heart rate of many babies during many labor, he realized that  the FHR of an unborn baby in trouble would often (but not always) become abnormal, and beat much faster or much slower, or become extremely erratic. Some of these babies died before the birth, other were born profoundly depressed and often died later.

With this understanding, Dr Von Winkel identified the signs of fetal distress, which included abnormal fetal heart rates (FHR):

  • tachycardia (a over 160);
  • bradycardia (under 100),
  • irregular heart rate, passage of meconium,
  • abnormal fetal movement (i.e. agonal spasms of a dying baby).

By the turn of the 20th century, maternal fever was recognized as a cause of fetal tachycardia; head compression and cord compression were also known causes of bradycardia, and hyperstimulated uterine activity associated with an abnormal FHR patterns and asphyxia.

When these abnormal heart rates were discovered very late in the labor, doctors could sometimes perform a forceps delivery to rescue the baby. By the late 19th century, the development of general anesthesia, germ theory of infectious disease (resulting the principles of asepsis and sterile surgical techniques) made Cesarean surgery safe enough to use during labor to rescue babies with extremely abnormal heart rates.

 

By the early 20th century — 50 years before fetal heart monitoring was electronically automated by a computerized machine — doctors, midwives and nurses were using a special stethoscope called a Pinard Horn to regularly listen to the heartbeat of the fetus during labor. Pinard fetal stethoscope was developed in the 1880s and in wide use in the 1950s.

The medical protocol for auscultation at that time was to count the number of fetal heart beats per minute (bpm) and record the findings on the mother’s chart. This was done on a hourly schedule in early labor, and repeated with increasing frequency as the laboring mother got closer to giving birth.

 

Helpful as this information was, there were still a few babies with apparently “normal” FHR that were born with unexpected and unexplained problems that prevented the baby from breathing on its own that resulted in death or severe handicaps.

At this early stage of development of fetal monitoring, auscultation was only used to count the heart rate for one minute (b.p.m, charted as a baseline for that period of time). However,  birth attendants and nurses paid no attention to the rhythm of the fetal heart or other elements now known to be equally, if not even more important — variability, the presence of reassuring accelerations and absence of pathological decelerations, as well as the depth, timing, and duration of any decelerations that are detected.

However, in 1922 Dr. Joseph De Lee, obstetrician, founder of the Chicago Lying-in Hospital and Chicago Maternity Center, invented, in conjunction with another obstetrician (Dr. David Hills) a hands-free fetal stethoscope for listening to fetal heart tones. This device became known as the DeLee-Hills Fetoscope.

Then Dr. DeLee developed the process and protocols for the intermittent auscultation (IA) of the fetal heart rate during labor, which included Dr. Von Winckel’s 1893 description of fetal distress listed above. Using his new fetoscope, the FHR was to be auscultated every 30 minutes during first stage of labor, every three or five minutes during second stage, and continuously if signs of fetal distress were seen as an indication for forceps delivery.

In 1924,  4th edition of Dr. DeLee’s obstetrical textbook Principles and Practice of Obstetrics includes a wonderful graphic that displays a fetal heart rate auscultated and charted in 5-second increments during a minute-long contraction.

It shows a normal baseline rate of 132 bpm at the beginning of the contraction (ranging from 120 to 144 bpm when computed in 5-second samplings) during the first 20 seconds. As pressure on the fetal head builds up, there is a head compression decel, and we can see the FHR go down from 144 to 108 bpm over a span of 20 second durning the middle of the contraction. As the contraction wanes and the uterus relaxes during the last 20 seconds, the unborn baby’s heart rate speeds back up to its pre-contracton normal baseline of 132.

If you read Dr. DeLee’s comments below the picture of the graph, he refers to a pattern of decelerations during the pushing stage that we now know to be head compression decels — the result of triggering the mammalian diving reflex that has been extensively studied in whales.

Dr. DeLee somewhat anticipated this explanation by referring to the observed effect of an abnormally slow heart rate of men working deep under water in a “caisson” (a big metal contraption used during the underwater building of bridges and similar construction projects).

DeLee’s work is the theoretical precursor to both IA and EFM, as it is the first instance of dealing with variations in the FHR during the period of auscultation by using 5-second samplings. Prior to this, the only information on fetal heart activity that was gathered was the sum total of beats over 60 seconds period, which was stated as beats per minute (bpm) and recorded as a baseline rate.

Dr. DeLee recognized a second-level source of valuable information that could be gathered by paying attention to the FHR changes that occured (or failed to occur) within the one- or two-minute period that the FHR were routinely monitored by the L&D nurse, doctor or midwife.

The theoretical basis developed by Dr. DeLee and its 5-second sampling protocol, which yielded much a more useful picture of fetal wellbeing. This critical aspect of IA  was widely not acknowledged or integrated into standard obstetrical practice until the development of electronic fetal monitoring equipment in the late 1950s and 60s.


If you are interested in learning how intermittent auscultations (IA) works, this link will take you to a subsequent stand-alone post on IA   

Note: The historical background material on non-electronic fetal monitoring is repeated. To skip that part,  keep scrolling down until you see this photo and the heading:”How IA Works“.


The invention of electronic automated fetal monitors and its displacement of IA

With the clarity of hindsight, we now know that just recording the one-minute fetal baseline rate was not sufficient to identify subtle indicators of fetal distress.

In the decades before and after the ideas expressed in Dr. DeLee’s 1924 textbook,  untold numbers of unborn babies unexpectedly ‘crashed’ during labor or were born severely depressed.

It’s not surprising that American obstetricians and worried parents continued to wonder if better information on the biological status of the fetus — a way to know what the fetal heart rate was during every minute of the entire labor & birth — would allow them identify unborn babies in trouble in time to rescue them before damage was done.

The very first attempt to record a FHR by using an electric machine came in 1906, which was when Dr. Cremer invented the first ‘fetal monitor’, which  was an electro-cardiogram that used abdominal and intra-vaginal electrical leads. (!)

Other physician-inventors attempted to determine fetal status using a microphone to magnify the auscultated FHR and electronic phono-cardiography to record it. They ultimately decided that such devices could not provide the kind of consistent results that would allow them and other obstetricians to identify early indications of fetal distress.

But in 1958, Dr. Edward Hon, the pioneer of modern EFM, developed the first system for capturing continuously the fetal ECG and coined the terms early, late, and variable decelerations. I don’t know for sure that Dr. Hon knew about Dr. DeLee’s observations in the 1920s, but whether he did or not, he applied Dr. De Lee’s theories and methods.

In just five years (1964), Dr. Callagan had developed a commercially viable system for capturing the FHR with Doppler technology.

By the late 1960s, EFM systems by Dr. Hon were commercially available in the United States (1968). Other pioneers in these electronic systems include Dr. Hammacher in Germany who reduced noise-to-signal ratios and Dr. Caldeyro-Barcia from Uruguay, who was father of a number of fetal monitoring terms, including Montevideo units and long-term and short-term variability.

By the early 1970s, electronic fetal monitors had been purchased by 20% of hospital obstetrical units. Today, 100% of labor and delivery units have electronic fetal monitors and over 90% of hospital births include the use of c-EFM, along with an epidural anesthesia rate over 80%.

A web article called How to read a EFM on VeryWellFamily.com concluded its description of EFM (obviously from the perspective of hospitals) by saying that c-EFM:

“… also allowed the monitoring to be done without one-on-one care at the bedside.”

EFM: Really good for business

By the early 2010s, more than 3,400 hospitals in the U.S. had purchased approximately 28,000 fetal monitors. According to an article in BusinessWire in 2012, this was initial investment of over $700 million dollars. The global fetal monitoring market is expected to reach $3.6 billion by 2022 according to the Global Fetal Monitoring Report compiled by Allied Market Research (AMR).

 

The report about this enormous market costs a whopping $4,000, and that is only for a single person or company to read. As reported by AMR, the business in developing countries is a particular plum, very lucrative in just the sheer number of hospitals and to be perpetuated by ‘repeated business’, making it the equivalent of a ‘futures market‘ for electronic fetal monitoring devices.
Hospitals spend much more on electronic monitoring systems than handheld Dopplers. This is consistent with a decision NOT to hire enough nursing staff to do one-on-one intrapartum care that includes the use of IA.

 

Obviously hospital administrators did the math and quickly realized that making $400 each hour from insurances companies for each electronically monitored labor patient was far more profitable than losing $30-$45 each hour for each nurse’s salary who would instead monitor (the verb) the unborn baby using IA.
But the human and economic costs of c-EFM are so high: multiple obstetrical interventions and painful invasive procedures during labor, increased C-section and operative delivery rate  and its many complications:
  • blood transfusions
  • emergency hysterectomy
  • postpartum infections
  • rehospitalizations
  • all the downstream emergencies:
    • placenta percreta in future pregnancies that requires a Cesarean hysterectomy
    • ICU admission for as long as to 20 days
    • maternal death for 7 to 1o out of 100 women.

The routine use of electronic fetal monitoring as the standard of care for healthy women with normal pregnancies is high-tech, high-cost, non-evidence-based care.  [EvidenceBasedBirth.com]

Since the first EFM was purchased by the first hospital, the push to expand c-EFM has been ever up, up, upward, irrespective of the science, damn the torpedoes, full speed ahead! There is no doubt that a constant pitch for buying the latest hot new upgrade is being driven by the EFM manufactures, for whom the sale of EFM equipment is so profitable.

Interestingly enough, the economic goals of EFM manufactures fit perfectly with those of hospitals, for whom billing insurance carriers and the federal *Medicaid program hundreds of dollars an hour makes the use of c-EFM into one of their favorite “cash cows”.

As for the health insurance industry, they can afford to reimburse hospitals at this extraordinary rate because it’s just a pass-through for premiums paid by their customers. Anytime the insurance carriers aren’t making as much profit as they want, they can hike premiums up to the moon and back and there is NOTHING we, the public, can do — we either pay or become uninsured.  So “cost containment” is not a necessary aspect of health insurance as an unregulated industry.

[**Medicaid pays for half of all births in the US]

For hospitals, this profit-making scenario is the polar opposite of having to spend hundreds an hour to hire enough L&D nurses to provide one-on-one care (i.e. to ‘midwife’ as an active verb) each patient, using IA to monitor the fetus as an eyes-on, hands-on process performed in real time by a real person.

Central Fetal Monitoring: The newest kid on the block half a century later

The introduction of central monitoring allows hospitals to centralization the fetal monitoring function of their L&D units. This new system for transmits the data from individual EFM machines to a central display system;  it is just the latest iteration of this fever pitch to replace people with electronics.

Central fetal monitoring frees up nurses so they no longer have to be personally present in the mother’s labor room. They can skip the patient’s labor room altogether while they monitor several labor patients at one time from the comfort of their chair at the central nurses’ station in the hallway.

My first experience with central monitoring was a few years ago after I transferred 3 midwifery clients with minor complications to three mid-sized community hospitals
Every time I left the labor room to use the public bathroom or go out for a meal, the nurses on duty were sitting or wandering around the nurses’ station. I thought it odd that a L&D unit with 10 labor beds could have 12 staff nurses & one OB hospitalist all sitting at the nurses stations nearly all the time. So i ask a friendly nurse what was going on and she explain their central monitoring system and how they were all watching banks of fetal monitor screens.

 

While this may be pleasant for the L&D staff, central fetal monitoring is expensive to set up and maintain, and has not been shown to be of benefit in comparison to bedside EFM (Withiam-Leitch, Shelton, & Fleming, 2006).
In another study that compared central fetal monitoring with no central monitoring, (Weiss, Balducci, Reed, Klasko, & Rust, 1997), there was a statistically-significant increase in cesareans and operative vaginal births for non-reassuring fetal heart rates specifically associated with central monitoring.

 

An interesting factoid: the 2nd most frequent diagnosis for Cesarean surgery in a first-time mother is (drumroll please!) a non-reassuring fetal heart rate!
  • The guidelines for healthy women with uncomplicated pregnancies do not recommend continuous monitoring.

Trying to build a better mousetrap instead of building a better foundation

 

During the 1970s, EFM became the de facto standard for obstetrical care in the US. and replacing simple periodic auscultation (IA) with continuous electronic fetal monitoring. As noted in the many studies quoted above, the universal use of EFM did not turn out to be the answer to the obstetrician’s prayer as everyone hoped.

But this disappointment did nothing to dislodge obstetrical faith in electronic monitoring systems. The profession spent its a huge proportion of its resources trying to improve and refine EFM’s electronic circuitry and making other tweaks the machinery, alway anticipating that the perfected ‘magic bullet’ was just around corner.

Unfortunately, this blind faith in the ability of c-EFM to eliminate newborn neurological complications created a series of problems, not the least of which was a malpractice nightmare for obstetricians. Having extolled the virtues of obstetrical care augmented by c-EFM and Cesareans as a virtual guarantee of a ‘perfect’ baby every time actually set up a situation in which obstetricians got sued every time they didn’t deliver “the goods” — i.e. a perfect baby.

The annual premiums for an obstetrician’s professional liability insurance skyrocketed from a few thousands (5-8 K) to a ‘low’ of $85,000 and as much as $200,000. This is often a deal-breaker that keeps medical students from going into Ob-Gyn as a speciality, and results in early retirement of many practicing OBs who, having been sued one or more times, simply can’t stand the psychological strain or the economic cost.

 

But the bigger picture is even more of an issue for childbearing families and for society. The obstetrical profession’s fifty-year focus on refining and expanding the EFM system diverted attention away from studies that looked into other origins and explanations for the unexpected problems of cerebral palsy and neurologically-damaged babies subsequent to perfectly normal pregnancies and normal labors and births.

 

Nonetheless, a number of excellent studies have been done on these topics in the last decade or so. While its fifty years too late to have avoided this misadventure, these finding are still critically important. This research was able to determine that the great majority of adverse outcomes (92%) traced back to prenatal problems for the fetus that arose sometime during the pregnancy and and are related to conditions such as undetected maternal infection, prematurity, intrauterine growth-restriction and other fetal or maternal conditions that created metabolic acidemia for the unborn baby. No amount of fetal monitoring during labor could have prevented or changes these outcomes.

 

In addition, these researchers concluded that development of cerebral palsy and other forms of permanent neurological damage extremely rare during a normal labors. This held true even when some FHR abnormalities consistent with intrauterine hypoxia were recorded by the monitor tracing. The science identifying the biological origins of CP and newborn neurological damage found that metabolic acidemia (as differentiated from respiratory acidemia) was directly associated with a baby that developed CP. At the same time, researcher were able to determine that brief to moderate episodes of hypoxia (inadequate oxygenation at a cellular level) were not associated with increased rates of CP and newborn encephalopathy.

Why the subtle distinction btw acidemia & hypoxia matters

 

The distinction btw metabolic acidosis and labor-related hypoxia is subtle even for highly-trained medical professionals. But this is critically important, as the reasoning behind c-EFM was the assumption that fetal hypoxia during labor was the causative agent in cerebral palsy and similar neurological problems.

Continuous EFM was specifically developed to detect the slow development of hypoxia over long periods as the unborn baby was being exposed to less-than-adequate oxygenation (such as a placental insufficiency, post-mature pregnancy or maternal hypertension).

The other known cause for CP and neonatal encephalopathies are of course acute obstetrical emergencies, such as a placental abruption, cord prolapse or a life-threatening medical emergency for the mother. But again, these emergency have little or nothing to do with EFM, as they are usually diagnosed by other symptoms and monitoring the vital signs of the mother. No form of EFM, no matter how well ‘perfected’, can predict, prevent or treat these occurrences.

In acute intrapartum emergencies, anoxia (the most severe type of hypoxia) is terribly damaging to the unborn baby. But all and all, this type of intrapartum emergency is rare; we know because the use of c-EFM hasn’t eliminated them. However, the depressingly stable rate of CP hasn’t budged over the last 50 years in spite of c-EFM. Along with newer research, this tell us that pre-labor metabolic acidosis is the stubborn cause of 92% of all CP cases and other neurological problems not related to an intra-partum emergency.

 

At the risk of boring (or insulting) my readers, let me restate the fundamental purpose of fetal monitoring during labor (whether IA or EFM), which is to help the birth attendants or hospital staff distinguish between unborn babies that are well-oxygenated and neurologically intact and those that, for whatever reasons, are not.

 

For unborn babies that have an essentially normal baseline, variability, presence of reassuring accelerations and absence of pathological decelerations, the likelihood that such a fetus would suddenly ‘crash’ within the next 100 minutes is extraordinarily small unless there is an acute obstetrical complication, which we have already established is not prevented or treated by any system of fetal monitoring (the verbs or the machines).

 


If you are interested in learning how intermittent auscultations (IA) works, this link will take you to a stand-alone post on IA


Conclusion

As long as this essay is, I didn’t include many issues known to increase the problems associated with our highly-medicalized obstetric system.

The topic that got the short shift was the simple but profoundly important principles of physiological management. Physiologic methods aim to safely manage childbirth without depending on drugs like Pitocin to progress the labor, narcotics to manage pain, or surgical procedures to deliver the baby.

Instead of obstetrical interventions, midwives depend on the normal physiology of childbearing, which in turn depends on mothers being able to move around freely, receive one-on-one support, able to use upright positions and make right use of gravity, as well as non-drug methods of pain relief that include walking, hot showers, and submersion in a deep water tub.

Midwives believe in the innate ability of childbearing women to labor spontaneously,  push their babies out under their own power and properly take credit for this accomplishment.

As midwives, we do our part by fully present throughout active labor, eyes-on, ears-on, hands-on care and generous frequent encouragement. This includes but is not limited to monitoring the physical wellbeing of the mother, and intermittent auscultation (IA) of the fetus, to be as sure as possible that we would detect any problems for the unborn baby and take appropriate action.

Contrast this with allopathic medicine, which is officially defined as the use of drugs, surgery and ionizing radiation. American MDs are not interested in learning about ways to do things that don’t include the use of drugs or surgery. As a result, our medical schools don’t teach the physiological management of normal childbirth. Practicing obstetricians generally don’t know, don’t use, don’t like, don’t understand, don’t believe in and certainly don’t approved of physiologically-based care or practitioners (i.e. midwives) who use provide physiological management.

So it comes as no surprise that over 90% of women laboring in the obstetrical system are given epidural and are continually monitored. The blow-back from this creates a consistent set of other problems that require the same few interventions — IV Pitocin to speed up the labor, maternal oxygen mask as minor but non-reassuring changes begin to show up on EFM strip and far too often, the decision to do an emergent c-section based on a combination of slow or no progress and increasing signs of possible fetal distress as recorded by the EFM (i.e. non-reassuring fetal heart rate).

The Glaringly Obvious Effect of EFM — turning healthy women into  passive & bed-ridden maternity patients

What stick out most in my mind as I googled photos of electronic fetal monitors and laboring women as was how EFM turned healthy women into passive maternity patients, lying still and listless in bed, so as not to infer with working of the the machines they were hooked up.

Everything about the design of ‘modern’ obstetrical units and the ministrations of the  L&D  staff revolves around the hospital’s computer-based electronic fetal monitoring system. The nurses have to keep graph paper in the tray, unplug all the leads every time the mother has to go to the bathroom, and most time-consuming of all, repeatedly tweak the placement of the toco and monitor belts on the mother’s belly find the signal and be sure its being is recorded.

Everyone hopes that EFM tracing will keep the doctors and the hospital from being sued and the icy fear in the back of their mind that they might slip up and the EFM proof needed to prevent litigation can’t be found.

But the real tragedy is what c-EFM does to laboring women. They have been ‘disappeared’ from the system. Instead of healthy individuals experiencing the most important and joyous day of their lives, they have been dependably turned into faceless compliant patients doing their part to make the OB department into a profit center for the hospital.


Stand-alone version of the section on IA — principles and technical skills


Women disappearing into the obstetrical “system” — passive non-persons laying on their backs, covered by drapes while their “providers” peer out at them over surgical mask

Obstetrician using surgical scissors to cut an episiotomy

Obstetrical surgeons delivering the baby thru the Cesarean incision

Breech baby being extracted through the uterine incision after forceps were used to deliver its head

After the baby was delivered by C-section, the mother’s uterus was lifted out of her abdomen through the incision and placed on the outside of her body while the obstetrical surgeon sutures up the 4-inch incision into the uterus itself

 


Excerpt –> Violence Against Women in Health Care Institutions: an emerging problem ~  British Medical Journal /The Lancet, May 11, 2002

Other important forms of violence against women occur in reproductive health services and deserve more discussion than is possible in a short article.

These forms include excessive or inappropriate medical treatments in childbirth, such as doctors doing caesarean sections for reasons related to their social or work schedules or financial incentives or adhering to obstetric practices that are known to be unpleasant, sometimes harmful, and not evidence based, including shaving pubic hair, giving enemas, routine episiotomy, routine induction of labour and preventing women having companions in labour.

Shoulder Dystocia Rate Constant Despite Risk Factors – Medio-legal Implications Parity, C-Section, Birth Wt / Ob.Gyn.News, May 15, 2003, Vol 38, No 10;

Shoulder dystocia appears to occur at a constant rate, despite increased cesarean sections and variations in other risk factors, Dr. Michael Lucas reported at the annual meeting of the Society for Gynecologic Investigation.

“It seems counterintuitive,” said Dr. Lucas of the University of Texas, Houston, in an interview. “There’s this notion that if we manipulate the risk factors we should see a lower rate of shoulder dystocia, but that does not appear to be the case.”

His study of more than 12,650 births in two Houston hospitals showed a similar rate of shoulder dystocia, despite differences in the risk factors of parity, birth weight, cesarean delivery, and operative vaginal delivery between the two hospital populations. The findings could have implications in the defense of shoulder dystocia cases.

“The argument has always been that there are risk factors for shoulder dystocia, which the physician should have acted on,” he said. “Our data suggest this is not true. It may at least be argued that we can take a population with a much different rate of risk factors and have virtually the same rate of shoulder dystocia. This is important clinically, because it supports the notion that our options and our ability to avoid trauma with this complication are limited.”

Dr. Lucas researched the obstetric databases of an urban public hospital and a community teaching hospital … The community hospital had twice the rate of cesarean deliveries, fewer multiparous mothers, a lower operative vaginal delivery rate, and smaller babies, factors usually associated with lower rates of shoulder dystocia.

Still, both hospitals had virtually the same incidence of shoulder dystocia: 1.1% and 1.3% of vaginal births at the public and community hospital, respectively…The community hospital had a higher cesarean delivery rate than the public hospital (30% vs. 14%) and more babies weighing less than 4,000 g (59% vs. 41%).

Babies with shoulder dystocia tended to be smaller (3,844 g vs. 4,117 g) at the community hospital .. The operative vaginal delivery rate was higher at the public hospital than at the community hospital (11% vs. 8.5%).

… the high cesarean delivery rate, lower rate of multiparity, and lower birth weight at the community hospital were not associated with a reduced rate of shoulder dystocia.

“It seems intuitive to say if you avoid a vaginal delivery you can lessen the rate of shoulder dystocia, but this doesn’t appear to hold water,”Dr. Lucas said.

@@@@@@@@@@@@@@@@@@@@@

Stand-alone version of the section on IA — principles and technical skills

@@@@@@@@@@@@@@@@@@@@@

originally posted  June 2008, updated March 2019

Previous post:

Next post: